From 5719b747445af8f0ba6300667510dd427530fec5 Mon Sep 17 00:00:00 2001 From: slayerjain Date: Fri, 20 Nov 2026 13:25:08 +0530 Subject: [PATCH] docs(ui-capture): agent quickstart and walkthrough, and REST, MCP and CLI references The UI Testing section's agent path: an AI agent sets up UI capture and reads flows and journeys through the Keploy MCP server or the REST API. - ui-capture/quickstart-agent: connect an agent (OAuth, a personal access token, keploy login + keploy mcp-install), what it does, curl without MCP. - ui-capture/walkthrough-agent: every tool call and its answer, from a staging run against a throwaway app, IDs replaced. - ui-capture/agents: credentials, the lifecycle as MCP tools and REST routes, waits, errors, what an agent sees, what it cannot do yet. - ui-capture/rest-api, mcp-tools, cli: references, checked against api-server main and the released keploy binary. - running-keploy/ui-capture: the REST and MCP setup moves to those pages, with a "Where did a section go?" table. - ui-capture/errors: keploy ui --help and keploy ui run are not in a release build yet; the flags of keploy ui run are no longer listed. - NO_FLOWS: a page opened and left untouched still makes a one-step Navigate flow (flowengine extract.go, dedup.go); the overview and the REST state table no longer say it makes none. - Sidebar (agents joins the Guides category), llms.txt, and links in the dashboard walkthrough, privacy and sessions pages to the moved sections. Signed-off-by: slayerjain --- content/llms-root.txt | 6 + .../ui-capture/ui-capture-agent-callout.webp | Bin 0 -> 76356 bytes .../running-keploy/ui-capture.md | 109 +-- .../version-4.0.0/ui-capture/agents.md | 127 +++ .../version-4.0.0/ui-capture/cli.md | 97 +++ .../version-4.0.0/ui-capture/errors.md | 4 +- .../version-4.0.0/ui-capture/mcp-tools.md | 124 +++ .../version-4.0.0/ui-capture/privacy.md | 2 +- .../ui-capture/quickstart-agent.md | 176 +++++ .../version-4.0.0/ui-capture/rest-api.md | 737 ++++++++++++++++++ .../ui-capture/sessions-and-sampling.md | 2 +- .../ui-capture/walkthrough-agent.md | 671 ++++++++++++++++ .../ui-capture/walkthrough-dashboard.md | 6 +- .../version-4.0.0-sidebars.json | 16 +- 14 files changed, 1977 insertions(+), 100 deletions(-) create mode 100644 static/img/ui-capture/ui-capture-agent-callout.webp create mode 100644 versioned_docs/version-4.0.0/ui-capture/agents.md create mode 100644 versioned_docs/version-4.0.0/ui-capture/cli.md create mode 100644 versioned_docs/version-4.0.0/ui-capture/mcp-tools.md create mode 100644 versioned_docs/version-4.0.0/ui-capture/quickstart-agent.md create mode 100644 versioned_docs/version-4.0.0/ui-capture/rest-api.md create mode 100644 versioned_docs/version-4.0.0/ui-capture/walkthrough-agent.md diff --git a/content/llms-root.txt b/content/llms-root.txt index 13aa917f0..8cab4b770 100644 --- a/content/llms-root.txt +++ b/content/llms-root.txt @@ -59,6 +59,12 @@ Keploy (founded 2021) is a CNCF Landscape project and API testing platform with - [UI Testing overview](https://keploy.io/docs/running-keploy/ui-capture/): Record real user sessions in a web app with the capture SDK (@keploy/capture-sdk) and read each visit as UI flows and journeys in the dashboard, over the REST API, or from an AI agent through MCP - [Dashboard quickstart](https://keploy.io/docs/ui-capture/quickstart-dashboard/): Create an ingest key, add it to a sample React shop, and see your first flows - [From first visit to journeys](https://keploy.io/docs/ui-capture/walkthrough-dashboard/): Record five visits, read their flows and journeys, read the same data from a script, and clean up +- [AI agent quickstart](https://keploy.io/docs/ui-capture/quickstart-agent/): Connect Claude Code, Cursor, Codex or another MCP client to Keploy and have it set up UI capture in your frontend +- [An agent onboards a web app](https://keploy.io/docs/ui-capture/walkthrough-agent/): Every MCP tool call an agent makes to set up UI capture and read flows and journeys, with the answers +- [Operate UI capture from an AI agent](https://keploy.io/docs/ui-capture/agents/): Credentials, the lifecycle as MCP tools and REST routes, waits, errors, and what an agent sees +- [UI Testing REST API](https://keploy.io/docs/ui-capture/rest-api/): Each /client/v1 UI capture operation: ingest keys, capture status, flows, flow graph, journeys, impact, test runs +- [UI Testing MCP tools](https://keploy.io/docs/ui-capture/mcp-tools/): The Keploy MCP server's UI capture tools, their inputs and scopes, and how to call them +- [UI Testing CLI](https://keploy.io/docs/ui-capture/cli/): keploy login and keploy mcp-install for agents; keploy ui is not in a release build yet - [UI Capture errors and warnings](https://keploy.io/docs/ui-capture/errors/): Every KUI_E_ code the capture SDK reports, with its cause and fix - [Capture SDK 3.1.0 release notes](https://keploy.io/docs/ui-capture/capture-sdk-3-1-0-release-notes/): What changed in @keploy/capture-sdk 3.1.0 diff --git a/static/img/ui-capture/ui-capture-agent-callout.webp b/static/img/ui-capture/ui-capture-agent-callout.webp new file mode 100644 index 0000000000000000000000000000000000000000..e61c14087f11cc274cad75e84bce73d1d78e99f9 GIT binary patch literal 76356 zcmagFbC56Hwk=q;ZJWDn+qP}ncGa$`UAAr8+-2LgtzUoVbjR)cZlCu$^Ph~I5i?e< zoNJ6V#>`Zf5*O#C0S3|#6IN7LhXc6fpfXduH%+n~%8m&*|3p63_;C`#}iM0K|OO-2wl=e!W(=edhfD z?)ky{x&meZfFHn5z_`N0?IR(e?-cOzmIJuHO8XH3INU~bBYpz_fKR@D0f3Jh00{8D zvb)ze8c+e?{b~C_{8$|m`27>}{rai#+ILU5A7Jd)@TGDB=m0nX2!E13#{dR70XcoG z0K_AKX@LlU*$>CE!l8j#0Pc6z*XftQ9bpe4dO+6C;it||(gI+%&j}#)!}B%!aeKSl z1TgGN3i$Q&{=N7W@g;C8(4FU*hxxPm4f&D#4L|g)@v-{=Xfx;o2=*@f;eIo|GCuUZ z3;gxp@on(M{K~r*I2Jhl0RXN)LLTGq4etBK{1v_f11Nq9em1`tZ#6yz0DxEi1%TJ5 z-f_e`;0AUWN0apR30Fh6ne=0E`wQsN5C#2;IKGAAhM>aZTXz&u9OfqQYo#>07X1nH%VxnzrLglwZtEnu^D>%*jr6Q?Z9e`f z?G2O^yQrArZ`d-sVqLL<4Xc`2qL9ChikJq>2kZ}PSs?FNKnEI{3nH6Evq&GNJquc^ z(f23d_9V?%KF|3XE>7aQ%O2P54;#VL$e6(gW)3&nC%ym!8hh)yL~Rc2Ob(3S@4Pe^ zTpin@Rr#<}`7r)5g~>4ZnpUO3#%p8H6SaN_%x4{s-575A0+)je5U1#UGt_px#-=8` zAgUfAvHakGdjy2|?MA7%{Lp2v;`-r4G5(-FMBC(keue}NZbx90lM?3mg_bG_^`#(* zo#M5MS`Y0B+ktJL7D+-cy9&dbxOPU@Sh6C?1K|wj<#5!)K+ScgakI_p83-K zooaHBJ>7M;eP981fgIG^|IgE1mI*(W&`n}zMEmr$&IC<9zpIzThRhYZDVPo|dzm7| zwe7pI0pCWj%|Z+YA$BM`5TpVv+->_Yf+u>!mp1*?(FfA|JJzd?7cAG3)N1XCtR(^r zf;<>lo?hMM%;iZ;J4n>^+Z{MUZ5@WI+Fd{d-FU>fjR3r@jWgBjb^^N^8-#}q1Rznf z2PBw)7Yf8=vTJhjRTM(4pTCGN!1*O4Vg}*#vO~A12r*oxg9yWU5J>+Wp>yGv#IDG3 zcS67F+fI5CtgulV*@A|}et(xvmdVUTXJCkya_N(%AjR@|&&aHs`WH;iyYWRhLAHeZ zKfd^ERbfWV@}b|Kb~cKxCr^8Dl?@hMEgv%CC~c_N)~?UM0UtcT`2VqDpodJfavh-Z zllqA@JKiOF?OE|ivpM=1paAhj3YgP%;lUoVzyxieWkiCubgSMpC3~r|hLc=bQCFHb zb*s2cZntVoiCAi-N-mvgyRMV;KUsp!#`8+NG`iM$p!JWxKhtnuH5Hx7{VrsB$_S5* zalm@LY+&&4*~}J8-bq)R1~ZvWbbu+xzuDO)fxauJtO#DeAd(+Jf^sIum5KJBW1ni> zZ5P}xrfq*Yfs0n1C^PDSvb3O~Umf^+aUMa+uSXj=RxpeU^KxX_779x-do6~_+>2Zj z?SC-jY{Q=qGtUa$K@v6$ORu*~CqC{=N}yxGsSz9USBK1*?wz8MIZGm6CP796cc@wJ zHL1Mz81kxmW~pT@$EnI6A0uLbratJ99!%-fl|)Jt~=pmZ9RhpPs)uJVYlL3AeDnH9T<{T{E6Cm z)5tUhVP4k&m&F{{V5s;u;Z+~DStihXXL|hForzVr6{S|iV7fp&ZG5y%VNRXM`eGsc zjWxgz_MF{kco>OSg8}NfpmH|)Xx|iW6iLX`h^2;|UgDK9sWqxWE-y~OVP?@~Sh%oH z9zcq2E|V1e_eYh}<=j4Ur{u#kASDy`9h_YwrbF*7E|QX8;S)&B7paE=zV1)G9b6IM z_0Y>Kk$2%2%68$BXr0{#>pY}DBSg%s=@1abrs>g-2nS^WRVI z8)8KqmqB+Vl{k;Y&-2(*eB2Te>-p*BG|)j5g-<2Y&=Q;Gw)8VmvwILZ9w0aZgmfzO zd|IGN!8i4S5wHLnQ!{OwDfQwFgy?6eV%Z2n<-oKckOgPV^?07JI zqSL1EwXVmpprv(!&-F~-^-Sf;^{ZWRW7CYF)0Odm4whqE;aITo2gHkwoYHv@hoEwV z!qn}MfW^I#R~6w_ru{rlB&FJ6dIiYL)-_AOS?E0Qv$3AiXs{Z#Xpy+*;>4HHrajPd z0{o%OT&KhVA?LV%>2En1T{9i|xXC z=9h5uc`0ZVJy)3W=s?|N8O@;wMTV?{;)phI9)9;XQ`}f9YcMqfO3t|leDmCql;`F3 zAp1xnwGlWwGA+>6PHrBmPDA(2#9QVy{#f5@LM)q)F}D6%6Z4!SSU8b-N}P#{KQ;IK zVw)eS8<(mPctAm>t>dZML2uOqPo4`QW}Bj0Q@R4Zu2K&zu=&y1TAw^Ro@k)tzD4O~ zr&x`%aQMn5)gx{hY>r-ktKQwm?ufbB-S0Q3C$gTZ7EG(4zxzO@g^kTK$7qvTv?Rc0 zk8R_MuEFd)XIyt}%VeKS#D^zq(O7fYG|VG;k{zOBCW*x)LY>A`5v;YkvM|6u(f$mM z#%z&0e{l-E^r}2?GPVN+{SajtIis2Ge39m0@$vMnp1}i(SDFKD+mS&$rYgddpQ+rQ zLWpXnAVwJHXq&W1@c%TSDQ|JmHDgxS9r$*tR<)`IoUA!SB`*B=W#-7&Wyl zbb)bXf-J9mDDEPV=55yWy< zaEig=OcL>6Z%p|poHOr5MD1ed7*CU*+|Qdwh$!w5o}flOf+?aDSm?`2--kF5%7w%Ext8>gabgsp$MGYif6Vnzoy>}hT=6)fQL2f0oI=O0 z{#EEJoAl(^;Z{Q{YkEhZ{G}_2d6xC{QU1s!secabb`i^otSfZFYt-u*u^udhk(aSwR+tzY0TdJUgW}cADQ^5 z;32NMP9#G-*?UDNzxw08e5+?W=uX!4SVu3H=yRCyKbY*38LW(XQ^>D?(ilj+^{)I$ zC{k{rX-#g%9}4Jp3^YUZliYs2Uv4i}M+-Xpl{$!}+af-MJ1mv%Q+D7wf7efO{fFgk z(w#GbV!dyd?0Ys(Z;*TK>|;IORC1?GxAZgS2A!}e+Ny5Cuk_!c;-C6{UM%oh=pdxg z1r@nY_CBBMKA$UFJAeNV>Hj-=P%J~;2FADkr(}x-3Jr@7`yw`W|2O`_(z~FBoo_jgeFV#)Pf~GmJ4hh!U7GQDR#idk?%?%TGhAK9K=3;)3jpN4N4& zW4+2RZ2A0BO1C|)QaKby^G75Q{vl&&4r7)(T}l_eDh6~-;Vmf34ZA^gLKvyD7@CH) zl}hR0^2WO&#apOgewX4VS*G{9 zVV_}krmt8e8WjIlRUH310OU@7+Sbv!!;UF(S{=+NqD@esRAQ~j&GYEasIIn8j8kif zr7khG>z*c~)4M16GGJ5&LPObig%9dgWc>#<&n z!27!zuk=nw<-Uvl?Po%@YqW|CQ!@ELC#w!C5t`V8RO0P1h932c8`RA{n6q@a^S~Jz zs<@=B9oc#j7^SDdHFPI%>l*p5U&u?#-?7@ydHb$@d-U6VCyD5L4~%hqAZ?opt+O%EJSH|U~pE$O_(2FbKU5(pXc{>{l@JzSN>m?K#r!24{RN( zw9Id#INd$6ImOq$1{=@#``_m-y9hC*{eg)b{^!2oFgb+ub|7KlLF+PP@XhKrz;dtJ zkg`%!Z2-7?^BINgaa1COR|4hwMOi-pQ8Pl?>tN>28o@@vTPI|wjg^zP_&R`UpiYk` zH+;T@`r6j@Q|QBw=oS}>no1Y0+Y=~h&eNLrg;z;5OvnTB;jSfpBrT5e!!5F%YTgM6 zk@>IWvrJn4ns0xbv$wnhr%+G1qfd<~Q8=%oE~@^wgOovoaHMqXc-JO*j8ZY$K!aI0 zZ6zAZYd>{6d-R`X-9t=j2dwr&iG~=5ivSgj53d5#ic|L?HBSbc!XuR{f~K|iN#%2b z@oyg_dm5aI@H{GEl!K6F)L5yXfKhjRQS8OyR61i9#DbZKx77{(q->$??+#7y2EI3M##M558-XsIxwX{% zh^5wP{rb1)iGtrHUZvue+J1_#Qao+#+#v~Km!usIGY<&-LdGTI>wJAJCDM?~#mv_E zy8+(w&Sx!L_RVp-{gcy9N#S{9BW;*|sV47cQ7$vChRq_2-z1cbO=v1cGl&KMQ%KFudV%maJN2*(5lgCV#LvYzr%B@*a zL*plQed#yMPQ(~~wcKNDvUMj|1SJInJKF{Xv{n9QShgzS6-OAAIgPkreHXaNuhx^* z6K@-c!G;Fl!vHl)v+3rIEn*CyLVhZ=lE!l3Py+CyU$;js=2bJ-)q(7`L;9BChOLT@VF=pxN){a)T#R}pZ zI?atO%3C!7*6*e$Y6Tw_jM;t9@8&e(YLgkr@p9 zqs7w{mxu+tf{UqAeC&a2R+v5w)#|Gm}6;{Fe8=_FiEg7zT6wr{IS;OXvqZxM2Q9IfD@ zG02^FT2%0#i2MI#R=;$st~piYknTw89);Ow5(y^bsA@3*$oo$QT<_YViDegJ`% z;G`{ten?lrl4Lgpy(RT&3WRb}!5B0`w#WOKgAC)jcyE;0xk+9b1nt;43hjNZpoIP~ z1(x+&yM7oIB{sXgKuFBb73;GZ!==UgR~HqfEu*(c#@*dhg#fKC1+N%-Ml6{e?(U}l zm=&%+5SY1cUq*Pu1AsTg^FSq9-@Ot0<&^YA4lMQS*Lgebj~^&p7%nrG8~YF z{8I$y<@y_+AcEb^pmrrB>gJ&XCXdnL57b?J8{x9qa=|&(s`^9;WoWFS&lWwiych`w z|1Fn_J@pzH7fM?%L#g{PWFXY13p9h~UQ2oK219 z&3en$+?VR>UNk1(do$s~?Hc>*8S@YKTrxWkKHh@m5|V)})xobXO5o#T`zLJz-HfRj z9h+Z9LdQTX{&KJJS*nG7bwHnzUo;8}Wl`mDG|gNAyzsK6Gni*nx5+khJ`R({+T*vQ zE9s9@84JDT-E+tglY>WiDzvb2mkC;gvAl|B_a%LKdS+_J*tDfNy~bi+!z{^dc$=B# z$*JsWwTH$A5X83wOE|Iirky{vcr+)Km@Ww*n7=L|3r^fI6DK5eC33tvftxr@9PFF7 zg(^7xQ${xq2*OVo)fK2;u^HQgNgyRKuaERHA zR))t7RrL7K(Jni#IXtSUpl%0;js%hQ@DzDBotPNj0h}U;?K%fnRs(U4LchJ={kuNX zm?8IcApUSEG^pp5BF)oiMD5 zO^gCLD^qt!eaw7_*F}_v-jem2+>|EO5*MhwwXrrKAd5SQMVE~RC*j$z4=%)=)$fvpec(GF7i41lq(~=ZDQZa*3Ti#Fe|D^z+$Fte7ZIBB zrA?9(H1oFvJ)e+n^H7<(eyg+}UpsQec+YwPrhM147=iu|MJkudlh?bDzUB0XZ>YC7T=SPLhv^%FtFU0*X4+R}y(#PNl3vFh9)7 zbFxnII2rc25y=>H31KTTmE{cN5Ri_OeXfuvB{x{xBqToiH{l>pMaw}C0}&n2YhkC@ zit7`8@#wMLC?*4T@YUU>r`w}{yE7zRLFk?f;N=-Ynb5?MV-oikDQT^mMB7^=eBx0(!vM_EK} zHLv|J{JT@)mVllM13FjXS;6Y8xeYeNCg?7kOl13Aih20z!PG{kOfj58N@mj$ku9n^ zUU>AVxn4no50vt6ujza*C7o%qONJn)>W?;z7WDt!W4SC7dX#VL<*~?0T$o<_fswgF z4d-|!WZAgeSU4W-MG;hk(TP6hc&iLzpFU)`dw7sO9uD!bfHo5vCj>tf(X|BwE^wHX zc=NJ3`wBO=@Fm4^^`3%|P%{5ib}aZGvhja1;Dtadi)$uR`VX*#vgt%X!krKNQQ<~Ddng!XE$8?-a>|;y8F|_ zQvr#lJ#wIKj%MkMDDALm_`1A;Ss9}gS6Gu|`5DC&ZtPHv>*f}FCxlx~62CIg&zrfS zLI`Oo2`QQM0}=v^C#&F>$(y+g?Eu7Df1Xh@8YdBvSb^{C9Q7oM zTS1n%G~6SQv!jj=X(j_LmI43^1^%X>?42f%a|mq<#L;w}S~(OiEEW7`hB&T)NXWbm z2MxNGAVeFKtf-t=2)rt|-t(E!R)V?mAAedx?2VBY<%gEX{ylminxx*Z*h}oae+3J? zvo0aF^-Oq?#4n+Wp6eYEDO!`q1~yQD7svpC`TY{8sJrIz6m5e#{qS_$<~B}6N-9Kn zAq!bAfygL)zl<2yG9ySd%bSzDO8N|Lz9$rjIc3Z@H7t?N5Csd$r=Ci_BbW^0>{-?9 zuLDHwz>Y%7&PzCFXnxisQ>?~R2Do_LJ?J5_p>N4H$R1E2$rdH{)V?#R^qq!t;5T=L zwR=>xV{P+t5=nEiv5?3a%3_4d<2nw=wbs9*mQ2QDh{%@a_@nGk-Em6KAypYbpz_bs zz?OBcNBLT=f-#CEPh7I8++mS?1*KD>sIZ^*eemAM8bRdbK+jTdJd7&s6WoNJjh)Rb zr}%PV1SBkoIl318_2U;*Jij7ZZzwLoyTqFb&?J<7ie<6)gkID(;=c50MRPLn4s_EI z;`F#Ugvl`y8$}0leyX@XiX1MAjHMk26J9n^6!kXPNQC(~_><)1_c02_y z`1K3}aYu`0@8cb_6$cxIj)v#*heRvQ@;3dC*cCrjKPsn@Hovir8bsLQ&}5mM9R=2I zCx;g$E|i9rKMwG8Q@ww9v4||ySy+aDZ4Z97(mw@$vbeH9Br`!PvIPk!^S8WM0^=&| z`-huz!Ez+xc{3U-&~bjLLm)WLlxM(UE!`{$sskTRw84_0R4|e(Cjjs?lDmUtBE)?I9iWqnm877e8hFs z)C%bDQ3{qCp?R46`n|e?T1N)6Xkssb4P4g^49rNS@^^K2dE<8SmcMk8gs(vH4QMi| zhsUHu5^qyQskGI=yCt)YRp{8?`CS+J_Dnsanqy@4 zgfw4^8qUCHMH0o25jH$(wA{94PE4-`C}*kZU7m-{aG{~GtbJXoc$d}# zq^kcURIDfEbjR<8F0bYHt7p%1Sj?eZL%X$74?Pz4e53fKd8G}OA7ZKe5h^s)f=!#g zAu&Q~fZTz{4;%#+fuoLEYY0(6`K)IB)^YSQBE`T()5ISom%4E#Fwe%Hdb}1nJ|-vm z%f2RAwlvEtDhHXmq3Z~h7+=fM3ZpuBHg4gCo89l$a|6#{x zw%2Mj%Dodc5SqprQ%r(#I_@}1D;0MxdXl{mRmBjESf2nB;$mDH@8dRdi1Xs`PdCl# zdV(A++lAc-=6z>6vSU4g2{DeF8WHK)A&i6VL@eW-#g{5^`zait&e9j<@r94%eeIpt z1_UNxUc4GwXHPFOWh$BOHkUClw4#Vhjx4DilH0gAfUYgHJ3RFEvtJUrihp@>-QSOA zxc~?0%H#EyVXx&miEiIPUSbcrDdx!$W+8GDQ)q{`SD-raU8%_!&Eu~7;6Jz(&?>nWnUKn{!% z*fVGe2uL0+oWw^v@%Xscf#du?+R}u-9XiYG?M15Pv)zh@4o?TmQ__KA?t+_~wbSc& zz!ar#CblN|V|C-sw666GY3p4Ko=_-(41M9OGv8ri6!NOfB;Xh$S^jLDHU6=yP{!B{ zPQb-MX%7U?BuD&)Vk>nVooJdwuRRz+X-fj$?Ij5l&@)^GUrj*>CD3;Y5~vr?KXuB| z_*+BPS7HdW&A!(A&b=ia*klSTPgSFfm`OuRB`6k%b`Ka$;#(x@WtlI3p5u2UH~4`D zX0zDn@L`-Set3#^WKrs44wxFq99du@Sy`Clx{ubfjMVZ+bXZZ@1}Hw58^Dn-)NnoE zFE#ifN4VciYOx`pg45DFGQ@vw+7OS4Uv8D3Y1*{^fi0)Ui^;vJ_i)e8GA+~V-Y#d8 zsGy=*wb#)~M=2U@lt$>9OhNj!B@aoQgj&tlYi(A-!kfMuEbL`zOHeaizapF}mh*Q* zE}zuyjpc_1Y0D+-U7dV$TAG4n9Z#tgN^wyTLTP zu%RqeKN<&AF3*Gk2}iNiD3SB3k+xg#W3J-`m{b{Z1{>R4pVXMsQ?{dKx!0L@K;>H} zcmS*%ALS4r)6u@@#o1Ekf~HhNCha=Qt3(=}k6ka<^^j{{5*0GjiwC6xNjqFehIe}a zx*Ifmn-pRkjiEZNHukT97s3H`mJh8<{39q%2AaFSk9{4&@oaMVTulax9R9#7t4!KI z{PvxMj0~O4dQEG#%nb3!qY^*RG7oucrk^D z-dpuDJ%0ZZvhiZ!BiFX^Cq4jLo{R0L3GBi5--&Tm6swy@2DNt&y@-#WvBg4xPrAik zf3C&9zH|bg_6=LVmXG1jN*Pm@-Xxh#WGQ4}2R)m3>o8)r`o)>dp|FVr0qDzSIQwE= zFJo#~ThopvCtvSF^?>19oH4#f)w{wh93!oWRg_JrMn+x8%-=!f;C;QEJiFke&N&+L zg^EO{?S>4l-64d;OazLKCfy2n?(D>6!1^ZRn)-G99+!Sa8K763C3X&_Zf4Gm?fKSh zAEbNDWG`$tNv$eAXF^zO-rn4F#}NOX-6lIJZ$?^If$e${6;Q7!$5SV^^SXKzahrL6 zD?1Q8*FN}HoKF<>laZ4VmfeNOQEb5`;7B6*FL;xlzC&T-rMObiqBlkF=1uq#&7Fd6 z#p%U81p$ZrmW+3h#CFkT`Cw=i|0zLiZf^B{$(KaWa14618Q#36)pqi4A7Ti~By6Y{HN`pxO9;EDMbb z4^_9(87IA-jv*F#eHfC_Toyk)vNAVHzY8HWC#j{iv^Af!%s4cxO$$NBB8!KxKUB)w z>5VqlrYz~$I%xC=rLV}-c{2U4?&$u7_ea?Z`bFluzJXsI4A$s>lyM(vA>d&_I83P? z^~0jq>M-+I!znA|c73>J6$OAWM;NJ)sJG+knww)gKrh}twWtL0*U-__WigFrGx%eP8LAjh039O*qcEiV%eDg9tbOk~v1eY$x)IHBVsHL<2|Z zSK6-h-%KlokD3#&on1z2#x+Y^?ftXQ^C-NK0pSUdtf@C`!;>)|*e^Qe8C}B4?$v>e z#f(*ULKD@*o5Ji7&^QQ40bl@Y=PS1O2z+VmnyUT?dq@Mj|0Pg#_O|(I%?n7o8s;vp z&r|`YNBQ|Xr(TTFJytZPr;4(ZL2^X|^N_bOs(K}q^pZ-jhN%aV5-+C?1_WiV)t!HH z{Jw}uO7Ip>Ik?*|?dlu47z~-&%168fvr9h0<*4GxB1X2J;yCah2J>kQ4}yO zX&<8Zn`{NaK0X5I>4$$>g~?*q=*c7-!rN_6glvRA%oo6k=BwaVO{2=nRFD$9c%-Lm z*`kiT*fEr+>wb4YsB6DY9*Zsvr9eUN%+$*Pt=Xr$&V}toWU~5Qdb16IE=3k&PipFT ztm3A~mx=32HYmR&NDZ?aN06n$1TVIxOqvHt2UXXNYQ#Az1T zMnS&NUMhT0+e$TRw4u#yfQ~t>4(yZr&s#!?8s^Qr zTwUEA?=PYgM(PUF`4zfjIOq$g5&v3456^*9>4sl`7p9CaW6rTCBetsnnC3VyN3f$4 ztxYkBot$_>c4r|>g{aG;u4qwCLB1gd>S?!JYt+GehR;dW9j(O4Xjzy+f=B+n^#+zK zJvg)keNqEfSV!JQs`% zr2q*?g1`CJ55RWdkZhNC_@))vo9B8Gx$E91G`@S$bb*(UVr`2`w<>tY7-RI>ymC9U zGf-Zl>^neAZ_r@z&K^`gSC6N2xpaK86@B`f*2lQ@$`+Bh%#G`Awx{;lNh$`K+eZB& z3L3Vu`+5i!!Em5A;?u)<99FDc>CP%ZzS?MjgO+GNW6U{CEf%N936X}F1gQ)|`I{>U zy}q<93PP^{^5X9fV(LI#Bg$ClI}nFe$Ycs|-FG)(Du@!$*bR6}FAx93;xtyGPbK0U zXr2nCY%55YF(2R9ev(jOn{lk7t3cKn%)u}f_4qlR>>Xlmf018k+5>!+33wO(A}qXp z+?j$9T6j`xNMaU-z*hn|5XbzM?$aaewY13e2Ex1uef0`i)IXCkfo^LS3kR6LQc3tNf+z6PXCaY1M$#PxaIGMwR2{%%VJ3l z84eHIx{uh$!OXSBTeK>WqfyA~%!9C{YTLRtmp(JU-rC9hvfnH;4u_7)mP^(aXEt-Y zPQPzL9gnPGX3~7?>}FtvfR7; zGxo*?0T8@2JdFnOcG!-^+DD?R`HE48xxS!@o^OxK`%-m0KL-7@*(ZiiyJdZdthsww z*qIZ2xXFl86Cgs#tVs1gX~b^Qlgw^=SxOIPj(O8yjZF8IAZ; zF4^<^wFG9HuJG>zEN&Y3GkiQ{dRYg<-B%kmFb>`#Ys=uD?Em~OKq$>+BlrPaZMrYO z;}vsj;`J5Ax){O#h<_z&*_;<&mT2E9H@&cGJnz&|CTf!`>IW3hfsrC^y}`3q7TjO2 zie#+Hq^$}0aCT;p=N&OvhC!f@+jiU&mZ0EoCY&n2jKz1Y%FSuUs1IFt^fvq(SIR4h zjly>G*8GSNJF1Ba4rEPzvHvpTf$Vg&x%(PB&+*ZelS(Cn5?Ias^ms!37E7BWxrnPE zB-$8VhC9FIbO$UAxp&<6ol(nQE_(7>bOl1>JY$Fl48J_p)i&RVQVSJ~*d+4yxiIjI zOJ6z)S;n+kEIo^foC@^}&&#{BlV&wkQ@tc0wa1?_x_({8RGSYu8*2dlnZMa|E?1*; z6_Nc9T?3;*%_ByUqMfohE>UNKBK~IaftHy@_zf<7i>|%Q8yIaG4)AzQsasM{n|*-$ z1}}C@j7A><93pRCYmbWf3U2Ns@q%f)Pv!$a-VmW+euq+fGpFNU`bA;Uk>FEweE`fg##WV0S2jSB40j&wS zF#_uL&W&KjM=QUk9Auslr3L&j>k8@Hp8zxp%!}_%D7YULK#P;GD4|Q}2uvG?^umpp z)Xk!E$$`6RD%#=FAvIa+97&EgBlYDtFkyS^2vo|fZuCe7h{FVc$Jo;BlQx8sj*-FX zwFQWOVyt(`wbP#AOCBX987$*cI`9!n2xA#QHNP-gqEz6C>BggkXG35a(fQ+au5YwAV$Ku)&ciT*ei`75?mN(LCXk3ICPZJ zB#=}5Xju)T&ZNV!t98OKgt63-8tG!H)-`nh-+!mV9RW-`!{Ra0P?}tG83b%Ss?ZLRUFE4 z1B-<$@FPE?^P@(GEj#{h1DjgE33s~Wx1^puwhZV>ni9@sdlF6_=E_7(6Wit`c#g=&9#B_!VV5cp9vv~wgylU_w(FZ>Vw!M}F?z|aX7Gq6h8 z)L0~|2ov3DMX&SfTZZ`UWgLW8dT%apFso|+ATRGg&kIn(#?h5_qXv53Xi#e<4!o|L z>LIEJ#uy)vYC9Q&^!1lzi7pwP>Ihh4HoWBi>eZ6%)>{__TEg3xD_ZnkMJ>bExI>7k zlfWvQU#d1jl$Tf6p0*W_`C?-Al7@NSLuXco{i+PWMth$hfylVM|X0 zW*+w~z~`O+v2JOs&iKMMoQo<*j1*4WAA3WhIjd_e=EOM8H=kic?Z9Mcn1XkWhBH$9 zwOR1ma3su$0Z#PKp1Slc&M+{~imb`U1ro~jG1H$SQp4b%`~TsvU`t+Ystl71XwGtjd26DOf3|96*g* zsC+6T`6a(Ts)+MQLCgL^?a*auS)p&bH_9T!d*Vd}Ud(46q84RB(zV(^jYu`BRR2kD z+KBBvLDWU}w`OhJk>gKjw4`P!@t`-b(P$UqnB-7hn&rO|$~e)eeJKJRxWs+WLSx4XAnk5FY+3E1fw@ApezzE{m-9xmInU3`^N>-5J}@Wz0Wig9%OG(yme zm!2k*oVO>vT9kPZV6+k6!JeN4x<{at;!8FiXlOg(TQ+z<-!d(fjWPj4PBeRE`a>97 zwe7OPEEv@{K9Q@0O?|uw-sw3KSj5lKvGfI5X4z+{Mn64ZGT(dDNH79cY z-nJsfL$u^ZYkGV@g6(YBADw6ZiUU`031P}%GnXoQCMkN3S!xka{gvv*JkK8S8NmQ7 z!Rf5g%MZhhmrTt1;~(XtOqwsdS6IoD<~$41txV#DCko4CQQJaM%vgn6YcV%~#AIDc zD#J4kdv1S+$lOaQd%&AJWuLM0;3D7zL;3l2{nd+ft-N>>x{SOnz_7Bj{tbT;sMmqX zToXOQktCvo2V8G{r~M6<3xoA`$eC(7c^5QGF-$y^(Vi9quUi5TS^leD-qsC++Jm*bKIKw{=Kws zHReQpSSUht45&0>0)$>@30PNQhipaR*iHxUdyNPf3j@1pZ+N~~mV|(p*Ic}Nboltl zw*y#|QhVKJsyj4g1S5dX*D8#BgFkC;XC_b82=OtQkF0;wSP zMOI(+COqxnf>=&VOL6xx?QbPOH+SFuD~$5y!@$C4NKipM=11cN;J*D&Z~NAmtXH== z#yR{Im5x@VvnkfEnYt3Fc#*X^M|0ibcivi zM$i~)`OuDZ#|yixZHG(Od-k{AMSc2z1OOnN_i)w~&en!odHeZNC~Bd#bA?gbNJ5W4 zohk8qhj>1zIk;)~@Ae0U=o(3$+zeOYq$fuJl zV0zgpmR4D1ItYPlVok-kWIMSF;JYXD&T)0&`eVQ@*|m6#L@lV_n!aFV-d;oFfTX7& zU-ykw?iel+1$t~ly1)Sg62HhQ6Rp_sYXrl~>}4ZiErH%GFD^r2}w?_9E*d1jsg0@SaH~qiUe($Y zWFP}v8b41tUbtx4qLRsfCnYdM;e1CmtT-B)<555!!lQgY%C16`1~X3`vqw(53?|O4 zJ(Bw&eMnJnalwL3`thMP#xUJ%bO|=|80Zi`THw*ezNKeiym`-0NY>lb9SgS9A@~zp zi83W7p@T1LA%FfVDQ9G}Yu+t9cNRNV5XucN?NUpG%X7s;&ib~^f3w)i??HYwagQMO z@}6h`Wy>~{&)$xhsorh>`^k$}Ar5{X;+iHcjLdmb`yfcd{VdAm)Yhr}BbiJ(EN{_Z z=3Hg;Y0%GGIA0ko=GQ_^To*i+avKpy@`XOfW2GB?9D(W2`(Acr5 z{xz;vAl-BM1t)EyUU96^K<&(iB-(mb!H}XK_mjTG@QVswyqeKkm|}qmuXt58&d0BA z9BmmLEk+*M#2nogyPyG?RUsHi_ZXH_l8{`Y5h4Fd0H4$q@6EYV=2?~HU{$kXf2nQv zM=KuQ^;}o%X)P4N=ZGRmPDx3!`2iIF*FoQ#*)?k9?8Ammb*>_49gB6BYIC7ZX%%A2 z1#>7z0Fn_jeng6K(nj>$ajP8xCUZKT2seR47dQxgmI2&a_V`s*wM(zL^_?2KD97>0 zK;$6FROzQs?5w)cY6g=g;wJY$e&-29fiGoA$-}VK*yyW6J zD3J{eoBm<0J**8Qk(pW8_v~m?YKaaM20s!)uoI0Kq8!Z z*}M{&=bl2a;5)!PENCY|#;P{2RUuGNT(UQa|9m*XzJ|N{8-@~XemN$Y^K@qV+L3)P* z#U~Ma9t8c{Do8X$8iVw-ehZ_0`hEm{ovP?(Ri?4HYWu)!$8(JKpWbyM{S&Yu`{`}B z2(4m&?ZQ2toV&#h?+mlh82j%tpl z($wkV-%Uz-2(_4kJ51iRM)X=E$>ey!qZod5n9S0~XMenWD@1qYw3>e8P@)^GDRz25 z->??zyo>vIBytjXANx=K2=WlU-3sK=+OfLH?2*4fh*b41AoEI^o*0sJmG1fiVPBAM zD~9yiKR>^_u94UbDj#x=B4HY|xVVE1)^RW@psL9l%|JE19~~vH2iMN~=Hbps442nN zy$_Da`c}leg^TKW?m!Q(iPX@2iS1G1N}$E?Z_&4rrF`n$Itxz&H{_Sk?F&5?tM`!? z1q0K9JH5;b^#5Nxy3IvnZ(drbu2+w{i6UYx z+#Rzb(^AbWL7)1apM)d+F#G-?72b7=1d^~z(q*p3} zqPz!@BsE*(7N?VN_|-W6UwT>8UbWjOz#Q zrRVMMuH2jWM}*GEz{Rd*Q-8(@s0}Wf^;VmP->I0RqVdROzsIJXUE2t_5N+R#{{av9 z-Kj91sP)_bh_wsFYf{?|f_vafPHRH-min38*Q$f2pOR-!-6#5BJ@fn}ij)T{tV;J? zB};!{&@)7=$&l^(a3uo|`zI=AVwJ^d>SGMm!;tZh2XweU_l<4kgSsnx1m^3;- z1#w`&v+n`X{6~0VTfXD^)wvX5eL%cxRn`m%_vx%4(pyFV|J_F$F|%bQ%i>(;hw0ww z(p{d4AZWi`X==4kB57)e&(L(#X35&zPl#-!dW{K(vQ-Ldp)&QI8}Gi9@S+J@I8}L&wg96`ES3dC~6N2bAVy1UoXnYVidja%@~cot?Qe zSz)jI-(&qG70164Hk5OdK_$Z#Di}(|QZ}3r$mqQ3=yuGYdJnW_ZO*5?-TxX z^Yhy0Qo*zog##vi9kb$R-9aaWnQ%y-n16T7`2h7?&hE4+6#{m0++n-zCU1>XwGs_a zV!9z|>y7j-CgBxMtP!ZeJHFY(EC6!I2K!JFK$~q1-Z^4$4b6OwLUoZ^HbN&vP<$!<;3mBfEb<+l${mrqcNQ=DCxmK~FKHxu zg0iq!@kXXYV)=c%8{T`QHbM&s*C!)iNaNqg%OXX}gTqMPw`RVSbC9|2L89~UAmf|u zC|Ak1e%yD)$gSYkNT|?$!f$l9`Pbos%kAXXIp}!j0`m86bV^A4vKv${4PZ62Wic%R zkqOf)DhxmwLKl{LroiDzm378{m=NVCtl=@V-x}_-fW^**FYItciO8~|VAH7}DevrH zyN+-|7_rJN1+`VEP=k!qk{*}4nr-ATjfi=8-I`*fpsjTq0jcT(h?vV|y=E;7l&c~` zD&{Pmq^%;mm|!r>Hvqd{`YetQiv7&ub>Ju*p++?K!K<8($jfxv(>?l#EPp1fTPHv_ z*(kDW8fMSxt^Z;-$X@0RKKoZveq=8LiTz+wkQ!yZ?IZ%sZLoW|WC*m;!V?8^U%(ps z8}XrC$tOZjW$@KAscH?A+NAbRb0EA@9057-E(B@#*uxRy%lOikIkJ5b3F&kX&u)qx z$_tEr!$L*VXmeH8}iuCVI4PSC@K{!g8@u%e4JVPnR92GtYa zM|sa?qd4-}I@{R#Yi-8P5{a3+LOU_k*x+BMozyYAU5nci}6oXLx_RDyn8Xo@-xG;98CUkB9BMBO48R^a&46U)?Cu&TC|-c z5x_#F3CN@gcYOWp(r9s$De{V%mp{WR!a=e7hvW^Ce$*Q&pEu(OScPx-cb9x|loNY_ zH_M((_g$X3q}9nJ_YOYM6o#V)gGB(r+w?Yg}~w>Pk% zpXUgoC~5WWVZ9%9()Wh&&^0<#=V_v|%j>aB>`e&+vUS60DjN&OU8s?{#@TrE98M^CAz4){>+gIwXVAlmEj#k4S6eU&*2-9PUSKHF8$E zU>2jD6&)>tUG*Wkam=!X;;>c2-;~WYfUe$%rC1U>WQ=zpL;{1*!=cOWxw_bD4=Yc9 z61nz`s2no{$Ra8Pp|~H9Aved0yPo*9tLy4RU3C8Nx_<5$;x&vV2yzw(BSw<4HYIB> z0mWolec9;2{}7d_)$Tm!`zqS#zyw&9z=f56%673b&X@+l=Gr=wXoHx0K$yL>e%6h( zn#kj03Bpmc@*BxG5Y%v)2u;vS^?8-cxr-0RbXxQeUfTp6jS5nf2{GZTF zqco&Qm=If&OabN!n-TXMFmvh;!Kob>zi3u2jQcv*ZOqfT!4rEo{sWP*O(*TVY7Ttt z6Ubg6u^SUj9jdwv`es}2eWT0hPfA*q&YvLXI+d0IK~wimb2pKwoUt3$c84F~C;ES& ze9W28ONw=oW$VH63x_wWZ!<>da{LkamDiOgR8vO6-@Wb5E}&3N7tFYmJFgK`Kp&k# zE3W@=g-LEhS&ztuVPN#{-q^H(e|VX~05y?2$vu4l0zj>Dtn56ukF!O^h8N z6@t9{kXiJvt7G}~hEH5x!=)oQpK7~O%w@1_9d z)_VX!hrqUo;M`e9#`q*>1acAVw(GTtCKW4uz{|nb=boAsCgAQxowyUJP2Uwf4s_O2>OxARY9Q zTY%D@JpSerB;baTy?+skj(?*r-#8}@6J60|OZW3CTy{}BW zseQy2Ob6^*bnN`S<%(blaM#{(i9eIyeLjPYWZ@46mFs*oUhrAuyC!f2j+MMI6+SeD zwZDaNS4ie>zh@5)Oz4u2vCS)NY!hm77%&p^qP9%x)VL|~TK(aNLjP*m^M363yNMT! zmz)Mqt@NZC!(?jG1#{FyFLD+}w63^Wz7E*V6cpvG$Q_z4WK=?XYbzd@HBqA&{b z3xa%%A>PIPJu`;NLq=8bFACe#eM$SD;@t@3H0#o`oZc$5ImXK0opLe(VeNZ$9tg;o z8_Rc+b-GsSzYt{e1v206Xmolq>V#okStx0cQh>6p*{>|XYBog2bI%|3AV15g$av=J z7Ri4&i$-h}pCkVC8{Mc4?OVeQuxo=si`kh>e)B8r4a945tR_O*|3xS1G+K>BVGS!d zyrRtnu$J^OmTZ#Vs6MySF~5@33e3WZTxzCwWAG(?#VcN`I!2oA)x*<3hcv z{^`u2<5s4>m5DL-HfxnZ6Z%9HdCk$?+Pbz`!os1DiGI0Y5X?4C$&=Lef%~MB^d3!s z$Dm}*X9V}r$>?U0zk)KhU#d$$T#>&dgWdO^blgv^JN6|t9=&&`_pBJ@WAx+>tWyZ@ zg;cG-nQ34J4GhuNW)|`l3^>j&c#zxlU^xNX$GO3R3{q0T>(pStFqFIJ*^9b0Y=&cn zSt5-7sWbgA6RIxZc8POdy50g<3t(bw3s}7CCO+L zt2^{-Q=Ml^$CF2(38akWhWuBqU((rd)~#QtIA+cj6!n%u@5>p%gU2T%0)<`4O)ni+ z*{7}tcDqpq(m@T8i)^ zaL4^SeBnC5Q(ADrY>OuTsut^Z)ZFj{O!?f$_uvResCdSGrjP%`7f#LpSV{l8FwPAJ z=>H7c>9)Dg=#P2AE>Mr;&{40@(jXfZUE@YV*w?->;eIIx;sEzqVN;KNgg<6uHb&M|fmSR7!v z1vXXqpqZ~ZX=d}|*@?38-^h59#w(P(&H!f;_jtRzpuNTu4O4Wkb%r!}PsUYE` zd@dF5;5&rBVtKqTV3T5{$W!M0w{^oSa(Gj;^lor-n^924-;gCo?g0>$NoCK~^4Rng z$nwTxPx&V7fZZq_KcJnSIZ$gxkjL2&Zm6}`RO>EHI&N1L`OT3|&7K0{KyQx%FcKUs z&dDA!M@~A;4t^ZJPb5nioBPuISrs^nwy}sI;P|+!Q=oa5`R{?7YG@ET(z9aZv~!iE zN4Mj{O%Ql8J%y+A{)#>3wtWn(EWxgOJ+aT!3MQL0?C_a?jopPf;TR7GX<%MS9d*RJV z{8|wrE2`OJS6FFz_*(s>*zjB_^HsuWqP@7cv@erN=g|*x{qRAc^*2Lfm00+p+?>IE z-)bkIkgTO_j2kqtuSF@Sa#FbRmRMtIsm)31s5!ycsy6ajh3Pjhgf-OSyEjDrXb;*S z2sBpk7sWxFEusjcJ#*i5`El0Nt^2dNKekZ-Cw0 z{hLG_C0+1^bPe!MfuzCK#oG~`^o>KCJ;zL5qNcl9>=Ef|*?(R1`)ZUT>3|nH&bC89 z|4>hknrrGpI}vUeOY&m}lp(?^6PKm7O8Q0Vu$E!NlD}3fI4(fZijMH7I+2+*4>z_7kPEETa61W4f$`1XmH0Q43d&ry zdn|o?w0qs&KsyCA7D3wP7}9<&s#n&G}Rg=*XF z{_lK;HqF*}>|hhH`s^jiGC$rVI%C|y2tT0VX6=jXW^Tsg>@)DN5{`2^RiK8Xz z8jUgGPU_UWhS^!fc^&?1_(`=DUK^>cD~9Qzgv(Iq%k*oUqCIYM6%+#d*VZ4y5dcSn z%OmzRVF5@qdQ>?#yIg=0^Bo2hsICg`7}g!sR> zMquU<=CCW24$|R;qWB%EtSEWb&{NbV!+7MdX$`zJMQ0M#?>xw6wQAiCOIg8D6YkI?$a0@Gl#MINsA)6>=SF~>%7Vd9%22Al~qkm`qLlDf*JnU zJrYRnPRL7pL^=gSTh9Fgpw;t;YG#BC|BdkRQq53Dy^v2TpsOMOL^Ep_e{&6)$%CoE zJ~s$XSDln+%-F4?>oIgdgdGNZhE^3YgUOjLeYsxM5_(r*xke)cw+hm3;x@GS>R!QL zM=v;47TbvmbMrmQz0L!{uZ>pEE<~2+zOOtWzaNGkA;AgjL9hG5x)l|iBH z|BvV3{PF)%i)%YMIAzMBGG+LM;1WmC*o25#T?|$4=)?bw{)34sDIB*=`qgAXL~8NT z9rxBA_u#vSM$H8b+HT0RntQgG&lC%bZU9tGSi3r=o1K%$p<5E#4m zJr21sAGpUF&iv~W5#VIZ$ZY{LA6}&pry&u~6M-imMXCEqc`GqG{jom2@=(dM8Fd)+UD#nVLb)g_F~N^dj*0ly3P zmCpAt0+TW`EfJc20Ac|3SEF6cH4Q;V(_H4QosY4ia7Us{=r}(I74U3(`9zsn9+}hu zvmUdHdJ*3lrqnT177%Rx*1x`?wn=E%mM@&B(mzaPQMgVAr;5L^tB|M+j4b z3%HF+ihr?hx&Kg^G)Twr8P@opRxzTajUzUdO^KW;1S7P15E}J940-cNPV0Ey1^KVX zo#V0)p8&L{KlEHwTa|zVc|_zSM(IE@l8(TXH0v(|ujMjISD1xYZ7HBx7`=G(nwEHN z8;8znGPPWSVn%Mxqb6?Vv7sM}+AFnSrZ*z>*4Dcrq+{1}e}6D3pmCX}%9g6fK8pGL zl!|gW>LDwkr5UY|my!gt*S1ACG;Xm+nieLg3>kdS4tAc|a6+mWMbBtr;M ze@rp*LgdN`;kOq~STwG$xI7IF5dHVsiQ7<&fqbl=f5dO1-&yS2)Ae2P(AiiJj?Qj$ z>h=zK$z#Os_E8Pmgj7$a;L!ZVV3ZQD>|o@NAr!il@Lt4>BAdE7Lmc|#{D#`T zZVAYKuDJEaFuja{atCZ%&KCJ2I)+?!gpRL4zGtczmlQD_eRaAwK+w70!Fo7e8Aeht z>phJ1SY#!XpV4`Xy#Ok&?(aEPn`@cH(G)!C^)^>`)EGhy4S)5JZ+JJtsk(V3dM){I z1E1=czF(=6aF>wnBBC*4Aj1SI&Ya+B82K=V4AMoOgDYomeLC0bm+8O#rKP$z!UlcF z62oa!-Gw4xG~WLdLj2;~A{vs|K_mU6iUPwCx=MjClW6U{KSpS!15Zt+ixPjSmST~x z3sn~m60jeLM)8xjyo%N1@Dln>c2sT}WeC}mm!S@g5WNU7UqK;_m9@<{%VFde#!~U5 zH`-lE$zut%v@??-@%jP&Z_E_rL#GTe3lZ*`7n(|~6+kAeJ(h;)$r}C=7kcfIw+i}F zBlN&@J|~f;U?ykOC{x-gd=;z${}DaWB=CYmCTOA3?d?A@?#nCMw!tqVRw>Uk zk5ggmC`$~?KO0O~uJ4(P8~Gt5Y?^u+KRp_w-swRhLuT9>hk4)1A<$N@Hzt|BJx-KW zv?#UD6N8`OS_yi3517#_1i4+E~g-`ee2z@IVRlm7=wehy`B>q|$)2fmwUYSQ49vGBXPQ~s)h zPq6mRdl+Esao}l{&1OUe)0%a@CFnsCVjec`y_tAf|ewr<$gyn~K+ zwHH7&V(FLK2n=n?><_5NLb-=ly#kfr-SJrKsk1U?h?NFe8KHMu@tUjpjPEMaR01@= z2OyPR8e>!Z%S}(RO$WkurG27TW~mghvEW__z+A}^X-7j~N#iV;OiC(c+nH|M1VsH) ziDMfLP{|^FVUMtEGb`o!edvG3!Un(jwLAKC4@y9q;8v|HDO3mb_G!316AKK7bdl-K zwgv;m^G9qp(53Sz`%ADnkLMK8{+=V09ZbK!J1I4pSei)R{TxC-g22Z(=W-3L!p8Ii z=0tmow6xaL*vJ|>fGZxKfNk|h_R0^NNOWL> zLDrVK!U>}33yq&-A&)t_fXZ>c;;JHEh-46{c@ z63%BVxTscnZw~cr`^^#0$MD1QK&13r^LGV2jdk!=8=++gX%xxYf+2URnEJPyZfUK* z0!K;r-{40;Cs074d(L-o7~1pjv5!$~adqk?(~+Vj@Y6`g4IQF0dE#~B0dTS>kDd(R zjCdM9sgxHa$)gYcI%lk4sU+=U`YV^1RgUzp;y?LRa_Semgzje|49=Jx(0PXYA}UUF z0T?4@%?9*I!5doK2jDd%uppa1Vc=+-`32RCGllqPi;$+*(L%RuKhY=&G#eCY+p;Xq7bv3?AEhHE7Fe`tcY z&*DXKVdfDKRJT8s&F2Y-`=!DK8;Y7-VU{Iw^@xa-&W2~~{#Nnolp^<(^~O_C*el)1 zv%&YETwN4y7gY0OOo`xlo1g#{w1x@%fOa%~yz?L`p4{c}&3xgs2wRK2unatEpIB8vcGCKY%G|wc4eUr|CR|Q|bBGj6eBscO z07NZLt#w6ihhqpH+L#94V%)m7M@PrfhPP%rx%+|eFBUOc1Q3a~9C`ZPah-?&Ql0b+ z_L}`kGW15M|JR?oc4{vcHr|38QsoM_9n7FB+78|{ltNioZTAmd^Ua~1v@TnT3bbD( zLwYP(>6)m_XUB?)s0dCu=^;q<8{ok}T^#@Lxs0NI)Z;KE{9HoZi7}~~^wk1kNFgc@ zD0xMpRDSY|_max8Llx6-q++$UQ8lZVzwHsRMDCAoN_0V^=7`*(tG~O@y^VWauk7eA z9--%OwEwhdp-v+4dG6y~o=eJG6J-%< zdx5k0B=WK_-#xA!;P;|3J>MS$^BNAh=XOvzMPG=gp1__72_sWl0002>&I+JK5Y&H} zZ}-Ym;&=QD2g4g?wA#J(%>TSTJ^)^t2T@zb#zro7xp>wyX z55^J@;jkks67BzoR{nG7tt8Jq6ojQT6Fb;p6gI~XfF#Nlv`EHEr{~-M|5>eAQ(pSB zgPAKAQ>Rk+ydcNIVCE~Xw1S}h&mc(sDg_F!4u9jMm=rGuNL{|Xl~ocO0AZBP-C!`o zjN1g&uoCVvq4P~=^3++czW-4iu1x}MLJL{dvL*RN06U1eV%Y>+;tGe#qr!-IZw{4$ zy9}FqH~xpEC$`BqDV^{{5=2=v>1XEs3I5ctnB0Ij+sgt!wKsw91Z_t=iIW0k&X+Zj zYIUFfW3{!{ayh55*iRWca@eA8N1(XW;&yox4zm++6VDY;g|0mdE>dq+q~YpoR_=VR zxG}jqUo?!tcA^y#0<#GWD>cC74u_tBAOg7oaE?Ow-X11SyrZL`_Wok?LCNrmH$d zkdLjB5k?t1t$kY+kuL8knfmi%^qtWH06_`gBAf+nSt>OsVco1Dl}(>pu7v_!fYy^4 zS3dcAljYBeV|#waj;2u7&{MWE)!i^q7ecF~#j|W8@?69Yf|<7*5b9;^-i2m}%;m{7 zK;S-fj)^2_$_Ew#ii%6T-0-)c3c zv7O;8>I#LELyR_p07lBes(fScC&jmLpD%=P-iLBu;z7jevrKLfi|D4@6PZ;^YY={geGiJcISdVFWfA$SbxFRMN%3XXLj zsX_a)1kSy)b`i|z`3rk?yK{*~H>JPyB}LZ+h#KdFV=96-7yWKKC0=0QYt|Nok@L+4^hFX0;`Z7g?whs;y|n zET`BCE`XU7jQg+*Cui4R@N@dpD|`?dX$7e!qJG9bv!zMR{QpvLX1 zBX^jg4}wooKCN)>;`>_yeac(((8My!kWP`X??Rj=S#NO@D{ees^*IJJ>RDtVFW^;D zzH4>az?pN0ADaD*A3-?%^LtcLa9`(^LvD`n4`cx6=5cxQ_|;_d;MmmCNj#bK1HMQ~bc`&-O5*jI z;Bz52PL~RNC3JOIu&jRr4#9hZ`yt(^n0yhf?lsgneYr zdLH;ny;#Cn_UNCRF#f&6?=Ni?s;-px7-}7dsy4NF%0OFQ{4kH zu9HI2<8<(}Xod*}u34H^N4F04>d+==lB}y&yOF#%Q1ox;+!U*SD!&BNIss(ZN?AvBxx@{bIOxVA59koG|Vj`<< zxS>|{+HTJ!Thgw*R0>XlcuBK$#>NYE-wLjIq@1ng$?7wbkv-Bas_6de7|;a3+yfh~xYG$LFZ|QO+{E);PNcOK75;UxXe+ z`BDCIb;zh9X|exAG|&Teo0uYe^Sk7kQ)lckj4lj|TeN1JjRtN)@3zCHJj32jWu}KV zhaL7WG*4iQzE-u=bI52uT*Z=+4yhkDd3shZ@%#Di>XOkj0Zcc*1rR^j7Qb7EuMe;@ z6sk^au7V#B;rT6kwhb_YETg(b>g%@9RfjL0JfBSBrH5O)PGi2ejwgY#r_zhy{H#{1 zls#)CbAkc}7Rrv9uTuF4`=J;?5sqYUMyOy2He6T=0xszuz%${KqzUeNoxIf3loADM z^a=g$cAvJ75mMR5><>8A(H>UdS$ZPXbk@#F+AnUX-aKM}&)Gq8=ps-#_J=#Ck?@Bq zfHe&F$@L~z1{|!=Ev`APw}YLsNrN1OUEh~mgI#8Ik7p(Y(SJJ%!i)1}4{y^>7`#T|o+WFa0-2@oBMp5t{FXmL^{59TfWlt=0GUTkEz-xU7q2-K*!~A{ zXlHzGN2MCo&sghs_iMC9(eD5WIcY_3D&*Q;omYfBwH%f*wJ-2~pA(}|mu*Ex#k0(| zHMZ8ZV9<<|@fB2_z&`a8Zy*-pqY7%b=(u~`Ms_%l4~_HHZi)(w_i$u#_p#bApP9Z3 z=K|FJJ{oh)1l`zgMeEr(H)`Fjc`Y-o7*jlX2+nR1;wo4S&Cz5%9~%sN;W(rL!ES;H z*z3{YXNg2$QC}uA(XTseC*wj-$SA{a$Qv1L^9XW;JyM1d6kQ!UASPNrCoy^dl)@m3 z%`8?aXKvkHRZ3($Bg3cw>uJh7H0Ho$o|N~S5ES$4zL=z`JPZJ+HtQD#y3*g!Z>|C| z35Alf0i;MyXf_Pv-I1o)Wdu zoFkjFfG$5u$?rLIao49#hLKI7yhQ`6Zr`0<`eA@E!@h%$bto%E6hZ+ah#6h7)620D`X*6sEM* zdT(v`HeSX22=~j#_kyt_+*)n)VIJiP@EcO)`1f z*k?m9cqXk!aQO5u_iZnsz}Ar+;J2XS*-a(v08#&Tg`aVVdD1GX=Id+`xEActdPs!( z6;+@rJ^om_I_<8Rwi$lCQ%LgF@U5LfLw1a5cE!uc70PwKWL|k+Mn!^)xkG`-vf*CVx}cS)`u1WAS(4 z1S@LeNX6^VD*c zD<BfG*YQ(fqn$7;6o6FaNL-W*SK(u_g* z;-8Y~%^vl#BjEAi?4B`==|cluHIP)mu<^tqe3gvXv^>etJ6(96fJh~r9V3C+82-#N z2+!rd3P()i8g9Savdn^tV82YlmQG%^LuZb63FYp;r$)421e1RaFWg7O0y;`BH#Jlf zYH=WeEwcKJf|HOM6`;b&Nj4Y(hki=FT)<##v#VZ6#c|E5OyX>j^r#B&F$LZI5$W{Z zz`3jz0LQQ(Kq1>p(D*0cs8vTJ=kRs(2onHL zpEy7VW4J3sA?!68uW`a)C|fb@#6Z#zqoL^I)9H#sIG8ld`90bXGqt+4$1L-WpC>7l zx6*R+)@XOH5umE#(S71c&}~rR+ViUI#}f|h2XLr*B9F_X?{-W3OG>3U^?eHb3_HBO z0%q$ichvpf(zupef37?$U_S#YH``wta$OLE03iT@bZd#nlS{^k#TcF{VV$RDn&DPU zE~63riv{ner>&#N_oxr3RQByC0wwN==(%B#7CFck?;LBzrr`Q@&0W%Q0-EvqYG&jGN5G!3cl8gf3Z42u8diBapsP4V>tatfaT zA*)R^wOTbGF5$CwhUt&PwrP3~VgD?4E*^($ZIwk%m&($i$Op)ckY*<}YdVKvJ{Nug zx}!ML;k0%4KeH-kakk4j#FuFXSlJ+e#4*6tVQ`rxgkEAuE*q%MN~P4QUH zKqU%Sq5o7i_!lHgqzH2;3*9kN4D;uKHt#tx&o==m_Sgc06o}aUjvI9LB6zXS{|Isw zGNwYkk-~VzOgR1J0@JAP9|nv^17ZKnV2t;wZQEnI za`CjB3gY&gpz9Qiq9r~Np25r(mQ;xiy;w(}$NRwPbzLh@s9$cM+KH&cSk&j>S)o3= z4?owGpa;L=>RQj2V#95gvONo zBUY(ujUII*05Rb9Fq|UQEd>~WS>@C-@Mo!QL@1C(vwD%^$in;DwQP7l5O)wBH`z10 zC&Eq0S$FVns7@Ju=YeHGk!rlanu5SY@2Q@A&FHV?65AuK(pqd(ua*0X`Ndo-LL#k> z8E9I=z-4}LzBSn1{l4@KwiCza7!qDI!=dAz9`-HUGNM=$-JJd$=~bcNY10Mz-l5OA zh!p|5J5S1xX?`XR4RTu(}G9;tC5PRu9U!KOz-%Z z-se~68V`}u<46nQ7e)XhCdboADFM>7f;CF@iAC>q-%?RZF)dGbHu3B;q+PBzEGX~6 z$BS*0H#Wh3hSzYb01oza{;Sv|Ewe~ZaU|NZ6=vpXaw{WFTtgsaV?B#-&4Bf9Y!iep z_P3RcXn`)+l|Te_9Oyd%DN zMn6M8ad8tT%?!IF(IunWMab?nzp&uyV>MatPbVWP59d64=->wq3!R(1CH&a`M?88X z@3T8SN}5nttZULs>}Ps$ZrZpd=+G&>vA^06GeUSFvmkughXRX@TpqUb2wYd~@==*bmsO4xvoRTJy zWM*g+4ljVgmg4UX>K-L5TFtn@Rr}mFBB`z`S$tunYTb>f@?dT0%|DB00dYiOGU3Vs zp#M7iAi&5JRi^MY>Dg{8Eot;o>_Mf9@3a?N57V95AV-=~*BUfiuxMLWCiJ@c?`cD7 z?Wa#ckEIsOcer)26d9nuniiovbhE6BT4>LApALEvcwE!--WF+&_LAO2pUE<6`Ue%F z%JboeQNKnwu_-UgAX_yQH-M)Qig+*nun_Z%H~20|V&Nt)If}Y(_9_!3A_RTyvVhvl z0@l$s3B@~jWjpRzsx~f#i+~2u1fYN(Z06U$Wf^6X@ODC#)G(Nlu00b7W{B8bnXxMa zkFlL1)drTQQg{Qi4-KM2kP%6d{%ilnO(<)+$l&hVcjF52fIdKjhM^mvodUn$sJ*e&Bqy%{Zg@|S!?B`0CRxS+X z==i%u)#U%Qf$kGVVU$L{NI1-GCzf{k=5HU4Dqd`-Wo0M@qfC_H+nXAst5I3tP1(oj zh!q}|K|~E3(FcBVw9Eu~-zke9D^hz|37a8IJsSmKjmfks_k5@`)9D!BP2}Tsy@9SC zX?Zoai2(7rJ}qh5Qu;8ynaF=9I*7}Q;0QPl6x_3CKqQB@NOGK|&>9Rp<8T_ok}2q~ zRC()eEXRmR^X_{8NG@#uz()jHusE;+};-j zZzD&b1&B7PaQaxpv+e8++HVqTX4*V$JXaWP&>D?z>Xy_pCmw%q75;ecdm1e+w0IZf z+mMpwpd-svh6e>OJw-TJwWp>P!wl%^Twqrpzpb=27V;<1@C>*)#*?8Ip5Rij)|iRV zR|ZnSwD+wQDLQS|3kRSt!3agsmW`AKTvBPb9KjaF*Z)2)zLtfC#~36weP=Cj=2vSK zi}GHp&>`$IO?E8?ycMwtiETGw9+Dyvip4Wl0WTO2l{gVQx$5boYC<%v#)5z1S<@13 z+(cx52l@sr)ZxHMeR=*JsDy)byQ&RU`aG(z->kL*CtTz_%!wx2HHgm=kg|>@*xt=7 zQ1LqLoao&Ac^q3s?qDDiH$(c!Xi>p(DzSxKb8kew_DzJ4HT!n>7~#=n(ra3wJ1W>e z$Wv6O!ZZS=coBL(W%D?uiWGB-0TLVRSyDqHXI6O$daeXv^-S6<_qK4KbcN| zi->0340D%ZC`JWEyX3}h%wP-Q zKD|syAZ0hNQ_52juzG;m7cGdO3enK>h0@iP@h+D!7Lhx1v%;?PXaJJ~_2n!w+O+eB z{)VyuJzJEhTt~*Dl8}CiHAK1YX?Mj!_@htjEP#vDr^~oA!6HWrB9Iz$wuKur^nvPo z;R2sV3^gc~9;78micz7d)VMh*v%K25ZTyH9w|s5|+s9KhcO--O;JVwoi+cRFeXYHt;knW$gf3qcJ;`{t;l*#C!Uo4C z#weZxEFgjIbpJ9?)R)f2La?cTQKq;iYUAL%@8>&*!{Ek~1~)AS|w!$lRt z|MK}~D^AEg*DeNkL=55JZdY`6I_^xEY0;l74(WNjF!2?z8eUm=GyYX?x}wnO?p%q( z*R!6{&3k)B@E3|`2z-eh1*jrAPg5bRq5ug?y&)oqwwv#~jRYd&?GA&hg)H-~CT zlr!{tRsD5%XiRy3ig1>}k!A1TSoW}C1*j`U0;OGd z@OuamAWIJG>u~n(AuhQBV}tu#2riu&OZqW~0s_ojv+I+a0kb5sNAsmu zPyZ?yPgr^`JLh#RmTC4bvpb1FB27&8!sOldC{FB~HZgXwI`{#L2-Sk^-Vp35{Qj9S zv$+>2OR=v4h zk#+(PSh^<6h^LKRA>C>FxO`PEN>(KVOUhdDY>kqs_GMj+W({%(uRdC*_e(pL6jX;S z@`mjDPw3Zx%~H(U>jm&c5(sLG#A4jbCrs28nI?~WrbJE#+yDBZ4ynl8u!W6@X;P>~ zesPt!)SE6V7UV}D*R^TDhtzkRzM8yFQ`1l$DzU(w8X+IOQx9e$fqk3sw8~Juw3Zkfg$!0TZGpcsXc(tRQxuJe!S{1MH7jU(>lLkF%u;z&D z6t?)H<1L=I^}#rVw7#f2h_Q?LH?nWm9vhKBn-oRpf^_JTteQlGMS>a<2WxkRQ$=UO zr%cE{+p$rOtG6l%ni;|fRN$SK9Tvx;R!>$CE294?rW$-N-T-3by2Msi*uM5kW%}O% zPVSnnzzUZCd`;3gw^@`wIYI&|al-~eYS&w3$HK4igg@!yyfRyps-Asv!SOFm8&Hm% z6koHn#Fws8e5yW;J?pN;m-M8{(z>kNsaOVFY7e&(O?2@@#6I~j0tJdsQ<3ENY9*V6 zPI|K!F2B>eG!i*Yan_RiNYd0G7UgZX8j^q5kD?`BoN(PFc`Ir1Nxd4`*aNSN4*VGmL*((!4OT;jWf1hz)% zPG{yT>O$zSiq1SH;3zT5k%@r98H9rp&zcL)gI|FG1>#-8(xE>%6+;&H-_|A~9r33d z=Z8dU-zxVw-Ewq~bhqD|^A;PqLCvega5=%7(4jF$vCnyB@oDj{ZZW=str<^Y(6wMF ziJE1UwNV7rHPBvGx(pkUb5vqT?SjldOgj%_hl8180nsh4jevD)ce8V5g1Wb16fha3 zk@xnZI_q-EV_w_TNFA(V^zZ+NsdI|XEZVwl zY}>YNS8V5tZQHhO+qP}nwv$RKtR%P2`LEt~YwK;d^)UAyWA?5Av_&_782yC_4w|AC z*cC%&i6WyyA1rrryd^IP39lzxY-0=Owl@!sl0@Y}a}+cSjx_f48oE|EGV{62y33Y? z7McAiMV`H#!S$=WlwPbQ$s*MC0mEg%QYQst4i0VDO=xe>9gCZI{Km>>?j})#k#*%$ z^_03Fd?e?|DW})7*y3gOa8;iztx5)lF#MuL6j*FQrTdZ=>?Pon;#?60?Q-3vg2dY; zcLBFe7WL3;DZ_J@H_C@4S94nb`(%Q)>y7I#5;~T5)ksGyOd~`Bs3@iqex=8_e7}?e zHo6yD!1fct6)MS=W?>iJ7KN>?$ViVG&%Lb7eCspTaVy>tlM+vOoDCh8O~7q6Yb_Mk5j$b0I`9Z58#ZWv8`pP?Krf zg^k~<#fp{{&OW$}Q=S09w1~IAV@GMmAsK#hR^P#wl%wE$u^(LlOM=nNSK`M0+caU~ zF3DQ^APb5yElmm?Ut`IsdoDoC#=)J;F^1jZ0-na4)z_EJr_RdAE z-yYZ!kC3Lwf+LXX<)xp?+#wX*m5UHteih4M%{W+Oh&9>wG4Fx1PPo7%>ER;XCAn(? z@EfDZ>UmeEM4&Hrpr}yN(vpIWW|{#${Hs8@{&b(7UL^JN%9Y$>yE%iJ+&R-kQka|+ z9k9tF%vR-39Ov=kGsbwdaNWY(xj_N_2iw!AZw!oKzN>NxBT4(qg z@Cd7!Uw+HXc^EgjA1y5wR0bI?io%CHQS@nq(L|{iISjqqN29tnMi4@3kS{QtVWP%w zn-6xF^=*ok2xo%X9&^pq>&}W6E`MtO_KwO#{)K-Pm#x@A2IzR3r^Z!+T4nOs6sA9> zd{TPI6NInT0jY|wG?a{8xF;fC6LQfI>6#h)Gz@}sDr8#?65g{u%E6Q z@+OPCq9#UaV>A)JcsF4~8@`c%cR7yGlxDU0GUUtJfJ%OO_ zDp^y{m((DpI|~vwo%UIi&vwPhG(8Nq1D7?*t7!j-QE~CvXHWKEadp2P#X*r0x%V8B z@%x6l^a=$Bk&Q<0ZJKdVVXtta!7>(6cbS1#?3a+X<8BqU(_gn{Iwv8&6z~QsnGE#j zJu+Pmh)IE1GJLOf?5zZe5S}K;@y8x3p1G{;{c8op#Is*JW$398hL4gu@==e%FI11|8}OKu>;+%id?#c zVGy(#DPghnlwqrp;xr;%>pXl-qM2YpuX27{9r8gld@Q4lrn&a|f8CJRd;4j{GYN25U9Om zU1!4ExODU%-jic$^n>no_p{ldEh!cOxyLL(P6C{45dlI)oyp5`7hkGQL6b6s?Ru^Pi2f7!kx z>X+iMvFjJL0wg06^^PF#fwnFg@$8|>a)wpJ>0ex!-tprj6K^VA5aDrezEob}%*YrZQd>ZNk(@PitN$8or1Sn}vcQt{aC5=%mY3E{< z7+K4ebWZ()p@F~}8{m}ACMRT($aY9U`SH-y$A*kns2WY zj?6@;dwI?JfbgQ=n$eZ}IG7p8gP@rN`5vMO5pwTTF7OVwd5iZC*D3P=LbQG6+|5IT zL;`ZbGkq_PG8(`rF*_4EKFTU9Q+Hq}a{uP2qUl>JRyjGi)f~F(ER{;fLs6xe*X^`+ z?J)x^behC8U*DpOZnAMaghZoUb~^Ri)ayV*93ZEhcIu7KeP2Lo&<)= zq=g8Mw2X799}aq7_QI6+C+KKVC;WRP8OI%mNB7~CD6Il4ps_u?0}_PEfb^f(*PqK} zbnT}7yPbeUPL(qz#sidve+{e!81ksCWZ}TM*)g+p5`5A~tyyYkGB~-yio6);l3zQM zPx_(zUM8tBbbPh9v;eY=QzgO(yb}@WFJGH+5~Fexa(j{^wqt}8vItc>v^RUi1WS^~ zDZ-N+Ik$X6=_AsqwK@N}gxy1hB1j8Y8X>k8KfbFaD`D8$1 zI7GMGpQB8}!-lFm*A`@HnBhh6$4E7}@GHkI-a|?z&Ku-ZIp%q`gdkEz0T6=tltQnn zyQOdS6fo$$kxet%u<~ev#E?v^YEeXPc>Z!Jg2Q=1%1)C=K+AS(*;##a*o!Bu8Uv+D z#ZLrB&NK4;j5l4S3N4P(FjEuTj}sFBfc24ixBHd%iQ>uQ;XT)+p9*3`SVTmxvdD7M zajdDPUVu3~J&n=H3sa>OwVoLl!izc>jlTAe|&tIun(G=N?B%B2oH;KyXRA4wq@#MjHa7L=1krAGIAW{@(CDh zNZ>7!zZh++d#A<4K8m&*li_1P~d{e zSN$_)Py}hpKzUIYh8_@qmPo6rkN|Qdj@a{6mxhm zg7RxS2J(36;oL|z4V>=W1zMiTXC3DK(jA+AM23m^M*VBAU2OKp;_~X$auiUl{gx$k zo8lG|b}GW@u;AY`OD|-!k8ToU*vUE|`jb^J)=q^aShq*SkC_2(S^qT6J&q$k@vX0p zH)@5xArN>4&QEv*WM{i1{paCB^Xr`Hc8BI*6jC{4AuamY?;z5L1gNyZHFd%jp2zFB zgW-;xiX<0Uv5Dq2yH2iF)C1EQh_Yc~?4&1J-b$`^0nJqZ->Bi5i$&AqNovb=NCjk@ zlK}9u3We2j;nBa3MAWtaIo9`WGz6_nCf#v1|Ls~H722gUxa5Ov6DCJ<-eYG|c}Hu~ zrXp(bwn|Xw%fw3`?xv=wngt-`BUK)n;QxL5tEs?^$k4{#sJRF`Vnw?54gZ+=*ot;P zjLw6KA4c2g)WZM8tA-o@&!-u65WNflVCf!VG987NIobSKiE4z7|6Pn*Wvk64~4oo*{063v7 z{gjL`kw}_1?aK^fr=_cwA*kobd-88wS(3 z`4(MNyGMH|L8)kcvR}(6%#CH2ds4-7VxPbKCUL}EdTB3zZgwFeJZP`Tf`RP5L~IWl=^9yNugaatJAZy*O*%Vw0;oi z(|ApARV(x5()D3LESS5kJ&VeXeqh@{j!?)TE#Ulx!|mlX)^u?7M>ujpBXHi9enBS( z&$=uykZQ8hH(D=@TTRqLaaFK|b0FUgeXC!BH{RydncQ@?YPv+Emi(`alMR2I8rzXR z2=A;9{NcEkO$w75RLkxrdg3D|5K&jHKxK}K)Au5rL(ZJDKz0^PENjQ)bh{pQ1$An4 zcOF2{(ci@STcvKL_t`pC2!Coz27c|Ygb}FHEAwi&A@Bv#85V&M_zTh(jgN4S`LP31 zEQh8#)MU&)MF*yVj!O;^ea+Xjm2>bP*RHd-3aW1yK`&LIMcKCj6#=0>Sp#AeW%f@d z=LV?sP)J?vh1s4mrvBb_{!|X!3qsV#xk4Pa$RASs%D(LQ3=^ATn?uWlE)IBJXHIOT zNcnv&4S4{e!RdlX_q!2TB0+GsoHuKb4_2Yd7Ag}&p+5!Wr6f7uc`B+ZPhaKIX~uwn3z4emV_5C1R0WqtoQY1;-T@S`Vn<%1?k=}OnkGWeGIpkw;C z#at&jcf8=x;eSOH#Dd64zd$tTrhCd;*3br@>o*eA9I3Qro>-qH4!)DCnc{Z!fi$9Q zHOb(QCF{Y@egQ57Tq30z^oKE8z^Usn5JL%$JzmhyGVmy%qwk-W6bN+SroHkI^knLP z=3Sh=dpn#RZR`AL#s51J|9=l4>MW7Iyed(l75AVNdVa*Eyo935|N9(R!;SiVhQ?q1 zmmo#o9!?^7I|l!{%)tMgg6II1uK}9>vtP>Npm!)B%G_cX@_po!C;`!#9+8!GWrsH) zHgi&0ujGmdQ1}mLRAbiJTrxL+KLs}*f%J%+ zlr{SyI&mBL_V!CHYjI9t?fTENUm6oeJFvlV(iXS)(h3$*rG9<#x99CvZ}NC2(#!9e zt?poF2;75&=lAvNk6O}8Kp^x@sU^_iNY~$)jFy6sv~|JfQ-pVA$>1i{I4U2WVS!B- zi-M7irnbarc2R5bxFV_WyBE6N3=v|Hs1=Idxdo8(F~L^@2W9+4pkIqKL$V}&yALR> zx1-_OCrlHpIa@7VkH>oDwpGvqE{cW(8=QnoUmj1JP@Z0l01WHybPycAk_Wk{8J@zV z?C(E!Vk_N;P!v7Xuuj5EarOxJL_&2v=zJ(enkut6Zf~Hra4AP}oME)Pb~H^m-Fmy* zCS~CNNUOgqx^n0B;ylHnh2ClS6I>w;-!4%z(hZ$Lvc581q0Sb7uXL_YY7dSj#yyIH z;@1mo;R3L9YqoUwv9WcwYn53jaR!)ORaIwoVCEjoZEb^4EBivx7A$t0Qvc}*-Ix-# zqE-jtYXGk=$btVd9eSndVGhBJKgQ*65(b_-odUHa%RN;L<--24pduwN600~){Fj0i zh*vZyy%Bh(S+gm?udX@$H@aV~^- z$FGYdsB`8?0h-TUQOI5JVo^IDD~H=!zG!I7J^;)Yeh+29@y}vGNkwjyVt9K=^a89t z(^$ov1|dSk4aTbzBE+D{`*O_vh1+L-NYh^l{@*QuhDfV;yH%n(xKZH2&O^ZERrTMLix>WromMT8giw$C|UrUUjTV!&(W?!sOW%~Ex1ocSEQ1m|+F)8s=aNirnX9nKJn72e1rJ}t-E#5TW=*%%s z((bu+jB%-X{at7Jjn0HHt8FePy7cw1jcP!00>vx554nU!Ek9io8!8SW!h3+Z$3Qg8 z-STfuo}pzw38XdJRF=@DbTsgf1M#(ax#q#!w;gDEfBrt-T~+7`a5i{x6=3eEQBXWi}xI6 zE-^D=5*^jtuTqTPpwo>9LN0ey4g|bAv-A~xCA~JmSVcvNl_y|bhArb0wrb!(mdP!#UqAKu(3zxrz-MFI ztnjSQWfq(xM4`X6>XlT?rL}G)Ob5co6|wIR3BQ2+TGuJrex-|V&%B%;WcNd}pDL{k zPRM6(Aes}~lIjZJBgb|w)O=up@{)doTtb9$x5+$B)%)ecNm{JH4jZ@%qk(dQI4yiHKKl{LxAS%q+L zqfy4^Y1-n>HmdRuri&R?{azbJ_-Ldw&C&92_lV8_jU5$`35F0+Nx79Tq;C!}@pIy{ ziHe!wm#95u~x4Q{FQ_b)hxcerCa)6xaz_>&;-5%92pr4$f zZ{RlF3}2E+r>BRAE&t@GL7l>8nA>9p|W%P3z3NrP5!~El}kL52y(;T zyuiI(s9ad(uSZ6vT7scsdfz3kmQyL=`p~X|874P zIxnqobIzHQrSsrka+ND|x{Q@B;V@QlQ>H^KX_Pl08D64y8pAM41860GRjJRtXFlG#zin6O8FhVsQ zztXMobcWWE5;sxKnR>S_M-np6DDte~&Ncn6yG@C0$aMsbK=W-(-ZjZ~n5HUmlwu7O z?_KL2VHUZTz~Cs*!d7#1U@hl;HvL=?Gm<$-&I!HJzG>~Fr{;)fB*D~_K!h1f(cjdz zcEm6~-wy?yQ2n!A@_{uNCE?0<7pNLL?cz!New zaXL|*D+KJ>M#!AVXqUZ=q@UGBDZ334Oa{eSW@dF*Ct!D}i3EJzmQ zrd-LM3VK*lCX9spG+TSRKfmH+XaB6&Pkh%34*bz&;I2Ig57U0UH(S9yekJ?#~xD6zVOOE>l7WuC1agm z8rN6c`d&zYYzIswMhsU%+6J|f4QWC6E5}aL=zo0bu-O(z3@t>GP9bEN(B(`OpIV5u z=*guSKR5>c2Hy#f7YLcajH4!eLgX77pw)4}i(=s@vWGMLh4SZaNU!y;6|fAAdp zW9wxbET7YSln*eQCkWSCui^ESodrw;Ak6H~%2V3=6=&jy;RNqZ5zl35rf~vbq`bRG zW!_f7Us04?RcK=U;2kL6h7rYaC7554T+vUpbA-&@64gMpu)!O+gX?v0Up{gJD!+d_ zb{_!FE7HCAZVhq~2G%2ezMbo$>c6%r%jm4t2U*L5!LEt5%M(qH=Kmo>TyGF ziqSj0?_N;>kvT`J;-VGLcj}eFwPi*n4}DlPzhb)JL;8DI2=A-fCoOxS9RqQ0h)_ZI ziZ!xs4%H#7tAGl1Bbe2#xtIMEUGaPk=_DSJ0zw^>o9Xi|QDVup=_G%)CVdVtRUU@e zV${y;EmU3$$W0sJC=6b-54mMLfPV_B1Jas9Ys#-c@C@gQ5s!Vyk*78aw&hY`m>sBd29^`8-4tp)+o8cNJxe3KGhJG$cT^7?HO8 zz1F+#nw|YhlW9%eAl+LPnXbT)mo9RbswO{OE#A3MrbcOa#4DcQD*u|HL@f(_h`guv z)8uOYlWHyA=gh*xq?FF^UCN`%vf{n#60RTn9@=Jk6$@0W4j zQ}u_R`$mca_UNyqY+25*Tp!bVK6q8_C?@K)(Qs;M8MRvVZXV;uV|>FH0!I$geGYi<*h zTW*-GlAne`r^e0uaV#w$jJ#Tjf9ev8T#Mel8y`>w7$&JDmR|I2af~Re>uy4UO($^K zs0&qrxMy(cSF9gbg@zrwAtkB}wnR-Hl_QCXN0H`X%?<=&r&-|_$h&E^ zo>GXw=qZV^aoav@JgZ~syenWox7FVCE~CA2{JKiQd9#ZWoqGei9!s|?Ud3u2#gPn< z9UID?V!ON5#qu90+_|VkeZNLs-^;cDd5#0ac>Uc1xt+#QB4-(%da}oRAe!A1#H@l~ z4%r*aM|&80qJYF$TViCyD96mIjtodDj>TqWziItf4cD^hNsD9${rvl36qV#j`ao$6 zsftEt-|#Te3_TPLXt^NOFG(4nZR_HuMb`TmL>@CETiES?u(qHTssrx59iwCr9GK*A zm`b5xBQSnu>EUz%#wt0T2*>$y%_kdKuizMpl8a0v&n|bXUnq!qykxY(yC3jY+T=(N z0zpdFji%|#L&?Wwd11e1y-R^B1Iq}^Kqio-jIbTD+rYa?zD+2!gU`q{(5gf;!QI_p zb*9Yp{-!OL=-H<9KtV!Y77S!LHo4MxJt~~S+<5S&YPMLHGF8g|D=N}LR^+O`a62IP z+rey_AL1D$usoFAGc3)h8X@Y z-%RF|{;0ZiytRj4M29%YPswKGRc=%u=et=2UzFMwT4>sja1cMfqv*l4JfXnI#b?pl z45ri46Pt!zWOGDYKgV`6v>!IT5Y6F%BWl&6 z*I6PX9`Y0s0~^4CAonNSdGBVDp$^0obU>)!;_F1c%aZ8r!-H3o3!^2dC+7G$29bke z)6Z*9F@8b@*T8a$q+jVQ_eQ-|iSc~M9x23#O>-OUo-dgK!&>RIxOgeEeCBEaVn)(& zZ00n#?L%SeU*tyb$e_p>DMIP)1e4E`MP93-OjyanVIV=xdH*|g$j7yox+``V zuC}l{SmWrY4Woeb?IJ8>8t)*{&Wk@JBx%|~euH#WO0)5G`$x1_8k^(UY#e+G_Lw!Z z92=vPp7)zwMtOkY#o1#eC!F*<(F6d~IB*eYmh|Yt;EQ!w3{{=%tlz~6@ZN9GQgAmf zx{&^Kx2DWWsn5>om-2UuBY9|?XqoDM_bzwb6unl}#GdKi4HTj8Z|>L8KOA7kX0!Zg z0|qOAqP!Ca;g4 zUsp5gEvZbQM2Zi}If?qIaE1j-ih4FV&9DWU8IGSD!G5eul3}^=#}Ok>1D}xIHDUH3 znd`L~>Ylg+G&52|T=w!aHHlwaW&6oF;$=odybVrS7Tn$Y&*oEZb82#qwV$v5Om>wE ziM+yUu$bVTdJ8-Q55gQcly8gPJx+`Ab-flw)%u<^%a)^M0e9ATyRtJ!TRa--c^L#2}57lX>Gvrp9YECo73GM z=Vr9t39YRIu2?Bb3U2C4iC}d#4#eT5_jCd*LNX4=RSFKU)`X-GEhfXdBqVwi8dtCC z8gW%;pO>88REC-uHCo=x&(cA2J1$WWyw(c#qz1E+UcZQsy%X`wI1o2L=Kr{cjg3yaARK4m|K6w=Y+(b+3`lZz2jsxmvmU9 zYMOd)@ou@53g6z1N-!AiIPx2aQBh=AaH~1!-AHq?i8nC-o9rbGb`*v#&6bEuGZou3 zJ#8(9E0{<>Fti-D29*>V1C_?&(Xxuc~w_+1@} zVIMmI?J5E?(L0MDBL{*r=s^I0K4{q)1+ zBAbeYdJZ3R*Hf87kwma`?Ea}1U;=Wj&(i!o9N9fR^J67$W6t#L2Z4qoncrWrXs zUWB@jzYk>s5O*+I%#)MC+MSf+Ge+vPQFzR<{z^A5@noIX=_)S+!bEFXeuKFR`xK2* z+9#|~B%3z8K=i#6#u~;NUK@)I#l(3`8sfxD{7g~qDr73~q2WmfdqVmJGgU+dPF&CU z{9Kxin)jFc0(cp|4yzrz%=@`;7-ig8`K(<`U|x$Ur&u83PM!KsM(d?AA-}}VC05Nw z<%&p?G3Yvwc8Li~hAs4?mGy^#&ESl6quZ=!amQC&Hudi6;$V}NJjJ9$}@_F z=d&G+++vt?;Ovso&zweF?6Gj0cP% zqgBr@4ZBUgkq^@a6a-m0KB27StuqpHcHvO)rGN8o&yc%D*I#ol%W56Qqk0+dsU}BH zMZd$G2?R9%jG0&2Q2-pgV)U~KV?iWFImUS8#7}F0vM1s)rb+-L%*yb`voO{$SoDGX zPc^$|aDgwT3`;6j&ZLODkpg&Rwd;9Guy#g%kFEyg47F)m3#oR^wekfcu_kx^0KYGy zbax;XctS?5udwAyrD4M7?Rzh2cK$~1^2p2;Ly#~z z$ODA}Qw48+pm~=1u>I7-@Do$>>HaxIZkrHnG^aO<%8#LJe+v1jDC20?h~Z4iN;~U+ z<3_vS{SGFBkEzUCbSEi!CHvVmXdQsTEPoz7E8J?)(O=at@z0>)N0z%$9R7 zE&8|+h+tQ57eTXkd7NA|CQjMc(@weUqczd#IB11yu+q<9K58OHt}|2&7t_`>rYh@E zN|C&D2Msl083_xEkG5Er?VD=*hGkJ5)tI58ef%?SJw8S2CPKLky7>qB%WRt)Fosv9 zMWV0kOn%@-ZERXvc;4xe5quz8VjdiP!ivFyM=K&oiWOC>ZhsK%crr)P)7W)BNzq-@N=(s)lpE-fc!l)JyIc6n^U938JrOqWFSgcsxTHJddn)OLbp%YB z(M+Cx9l*8uhKJgVouvj`)Ms3ORj7=39&w|-FdZv38$x!&8bcen}DHTLql zNI*{r;K1pnLeF2?P<<0#5pAlCBBB9n*Zver-_JM_b)w;z>tuNL=>NJQw+Q3M+867O z!w>?x)!k2uys1Gx4-o#*7Y$4?SwQQyTZWr_?^m}56w)&ygL7BCmm50L0zJa z%Myk7r5s-FZ4u5Y;KYmk^$SHjVi z1h5xbxdWX9OgB^xCE10L5R>gF3w07)#xWzkc^`P`eR?E>Tt?$RA4lI12HVbIvmmHS z3W_N3FyKnnZK%9LuiV~AQ}_LS?QVvm$ zrjJ-H(&YHV9fhgk4y(AJ{-fL!$sdd=h5cd--l)wN70f;~&!-5XTn z%}^DTvPr}$bvKEb+TC~0>2=xcnSm#S@|=le)CbyS1jJzo>;r_x*_#N~Yf(M;J28ut zQUi9+oF^XaZ5pax z<6E{-a$;l|I-M#gOk-L?3qv-^P>vrZFo(@EO^TfZWV|96xJDGF$9q59M(TY2=sD*V z&C2c|Ivzyr6P08n4O%zA)=}NW2Mp6!(48&^0%d8mjF(=R!Wr-n*2Ath-3Ka*P&cUF z2E?C|KoR}VG$5WKf{cbr3W(alEU*s@74f@ACmjBqGBqA;YjhFsnxsl5(aPlWPDHe4 zX!sZYBleg8n!4Rj@>*a@EY;FEe)GxC5~ z5B36~r-=Pc!R%0bQbpl@%;pX8sv9;QBhOrdK?|coH7V5isGx2{a%Vj;g*m8Szx@jS z(K{#&U>t|ef-byNgP=u5KoL9gXVdD#fb!P%z}G6VQDY5Y>}R~I`k$6NCPgz*Ci1GH zGOGhMH;qEo8);PB0gTk@^`q^@B8zxZ&DK^yZTnpEqo1o#wr4s&l3V+;l&dyQ4i(C9 zAudo$YlgKkW~|UWWEK&MKhp)-gkJyK%&Ywd+pv^L9j5b_A3W&ik>}0xFz z6n*7N(*xp12;3oRa;@EQ2@uCL8g~Gj zXi7dM7>c|8(^013QO3-?(!lu~nqGL4P%|La$~7KcO!*e4l^p8`1^QA+h6NT-k%w@O zTK&7>N6Hx&ZzpX3iNIy-$VNN3d>g_U#!JnT$ShjU;rY818^#3*bWO|{-yY@v$(=WE zr}0!Xt7&$$lXG*U{#zc&_*V!~0?zL@A)bL?c*_}Cc?FmwK|SU<;|6prshuSd8(bN- z7!SUkL0sohkY%F?EARWioAI5jM*JcPEYt6^ve9Qf(FM7P5hp)_pVXmp1E|MgB?gB2XcVBbh?u5% zl$`g;4ge?(u!CB#wft+nma;$u2qYL)8jtkbOz#CENi7G1y|Q5ZL59J7XTFkp;F64f!9UN`#}= znC^a;yrNa2Jz?T(YAIJ`BV$R3p7wx zG!Qrl&EC4z+D=mu{SJ}ugV1ysQxX2=6x5pw{{Djwhl?o#G+X`7YoDlaO1uo|)88SU zy}+xC4ccu?w&`y_)1)%q|bwGf^c&McKOnFMT zViFRHtI{y#9d^LRj4C8$ST<4TZ%Ch+_td6IX2BcisO`cVw|TK;BKZZd#CaQxnzwoM z6p4wO*e;Dwoi+_4(+Ce$QzrWFxksZQZkwM0ScMltuj`s;yWT+tlL57k;H~kd!MJZ4%tH`xDJdqT{ zkXvwOUd=r8pgdpN#v%x6DVM+(^LYi6l3Zy7D<^GwLC-v}pID*~r9;9-`6@t(6`7JA5&>V?KFl5k|FI>Xj1IqUk>BXIyvRrsGaV)d(QQm>z`F%AiV5unsA{ z8wqqBo4qjp-e;^FXBK(0|0%|2hSl-$OU=m-BWPoD4L{icMli@XpN`q`gGd3Fl~==! zU5Y}NbzvD+!*Gbq4^*Sd{o#ZIe18>a=||;S^zaB_xX|Jp?Ei?H|Q-n%H*ul zU;;)>OJQ>KVttKch;{F943oY+`wi9cHBsvK+IK$7sUl@bH>CZ{2L z_?T}1+41uac!z{ZWxHMJ-Fou$F8LZG6dcqr?g|!JUw4?wE+m3ge-R+@r}A8s>-Chp zEU-m;V#drbior}C9iz{ph^NLWulG0eeQ;J8ucto?e)h_(iGp^?DbS!xdrRv_T_%s- z(@FC8(B-IE7>k~2ROKdNCPZ-EGI9J8XIJ1nQjZs1m2#CKrcsX~ud>bn z80y1XCE6HN&eQ#OD(;_|b9T-(FC;(F-J{}4JC;56I`;tjAq5UeX5U(`i&^vY-fFTE06lCmpJ75Nuw3BP$$53AUoo^U$FfLzYNz+q}QczH}Ar&J94)@V_2%BzmBZv=!4da--B57bE zx+eMX;b7E=7px&yzsh;0L&RVoQ~w9iLvYx0!O^sCim@d*QS{x`3nDF3V3@~}KC1FI z3z72SUGZHHeT{o_icQ)0T3$RNOnPS?V4X8 z#GZ7%44Umip42Lv63@qI@lnzS?u>QQseD-8x!oJ4F1iK29_iJLLkr3UQ7v^dih0SR z*o1kDJJjKsXKUWu^DT^&x9_t^=pHk^Zkly#7ULgkl%}BdFSg3(+0(0=en#++n|VuM zE5GL^J3-4kZB(%SRHQ(VJ(vev+fF@3g%2^=-2sT$1_M)6GTF(n#33AN zf?m+-)O@{;WW7n z9pxLb{G|<`p$x1|OHu}w$BZ28*xySAy7MZ&oc}&pWR!o+{oV@lSt^)yK}%w;GUP9a zm=Tw5X@wK2lQU0OOi_uZyPqat6BlJaA++B#R?Rs{w?Rmdc4c%%E<09PszI(U_uD9cu2avKO#DnI!$el ze`)Qcyhb^4ruC~WU>mZXg*Z85boFIFuYSbvt&SKtKoR?)t&A#qTdGd-(83i9ZOt18 z#*E|RM`j^_xMiZbzRSb0QjBFZm~X!4nurBQ3+@6- zgxrH$4MU!DV2#*~P55+mOHElnLWD?2JDY$j_n^zpCYX=O_W_4^4-YA!(c?t69BLL7 zr%*q_6Qav5?FJCPDp)p>`_0vYztqr;Q8rJ8cvsx?)+)ZNA!5qw+Z6(m9z~J>3$a5b z^EhNhGpAZaPp`}T-df1{YNv8&P${mzcYrS}CKp;Vt>Yj(E@Ja+eL?owG2a>60Seg= zW!^>qWe-|&2>ewPN!;YBmSpp#!aW~qzGBa$7G%J4!g(){WYVId3Z8HSQg(84UsKr} zzO0W{^%&FrXaN{q=?>EC_t8W?lmPpe|f)A7HaO>BMQ0WVzEgy#Y6 zibz4#e|UqiXIvG3_3-F-iTJlUu z@!a@rYnjdxm7-4!)VWMHM%1wernx$SWMqwS@ZtditXVMRih^tJGo5HMdiSXCmCCm4 ze&^nbcAN+`PzfFppX{b=VPifG++%DB0`b&2C*XSwJ&+#3(&I5$3kV2MKbSS9;*m$6 zg74A4ThxnMLi@UY>(ke};I>}hMj>z$fL8F*N@mXVKoPxm%(_#l*T#j`Dm)|vyMiEv zq}8GCi>CZ7Skf<6+%BphodI#bMh(%BfYyisWmR-q?!$|KT=GEWtl#`!+sMd~15)^| z!#*+6*Zy_Iq-;MJAr}7HjTnPy@<1k50XD{8nrcf?0;4kr=BJFy-EV*YXj%{U7qBKtwuG)NDBHcI!!bvcN>J>&r&cjX2l zw)wSFGMp_p;ti9?lG;7vvo*xq2zUBVPWng^W7m!>anwFbo2T}OSl7XYGpgHn!YKH)>4yp_J( zGq5yZjCbfz9sCRmU;DR)o&eI~yOhkNiOIXJWt$#(_pjq>A{+M8Q~{Gl=qUScLBDDo zM|Y2R9&br{!5mkW|3%k11&abJYkJvhFWa_l+qP}nwr$(CZQHiZefK$Y=V2ZuZ|PKC zQk7I!y8r(JKCJ&B-P`12ydVm^ZLi|4CA>UZt*}rKS7@x>hFF9SL=6-`;s_cUG~DOc zI4=V-Ej3dj=j}lD+8x!608Yp_1A{JA{?W)ZhVL*<3i4@6+>^{fasTvsM{WN~e;l!1m=uIzd{{Gb&1rlor zh|vFizdZ|SSm76=#LbUWp6JcHnr<)Z6ejRv#m@OCTMf`CxKP=F7;ZX<>!b#;D&RiD z9<&QK)+Oq@n%U{W>M{!n0}I1tml@(8*M2b`t2fRP1qi7@lsdRFl$7U+^Z9M~M|8l1 z8^7zhVe~S<7w;xo8zEK{UbKLY_F+gxja(>FIN#D)xKoSerL?SgKI;;r`J1Hm#CxXvc!4Kl+IMY{T` z)ZlX;)oOV>d%6nEFz$Kt!~Gk2I)aOt%CXcP(+zxr%YV82?H>zcv)b3LP$YsHN0G!f zEqk|d&a&Q8%NkezyinerAkZBx&3NmV?=$hK2Cw@iud_x0G0NGu-^C84M!=E6aZX-aPArBm05ZFDjZgDe%HWb z5R1RMK59yI39fYPX<`aFq~b^Gh(8wz8hPvt1sS2oXw7GODLdAPY1O9Ry!4F6*Z9u5u3d0f-@(zPE%8^9oDC z2Yz5V;8?OM0cpVckRWgjzhT*5x}?*wQ3xRTu&z}>GIYm)Nr%&P*=KcPGi{xJs2A$; zoZNM5Tp`A7${N!@DHT;22`Ch2_i#SrauOmLnWeEQS(}@HJA+%Rj(uusiCsGX(WiGx z)MWBZs#XI*fihsP9Bz;5k#|_hvuhe=&+4uI#_cpdh8rIy$kL3Gir%S=^G8~aKq;; zPm9rk*@gCI0&np*DyS=L+n_J@^jGJrp1G0mjmImV{Gb70WDBK2YVSgHHC=QUdRfIY z^SQw0^WNr@JzPopymXklkvw^+|G?Ct;`gefWQ2sX@69Gs11p*cnnd|EJ>C#ucekZw ztAl+masP0rvMGUsW$PapBWlI=#t-S9a>MHj`es`i@WibP?<#iX#i47?2o{_Jqcuq* z;Xu?LI$wD9w^_xpBJ@a5&a)`s$?q!%T7DO)9xq^kS{7^olQ5&xfGz%3PXR7i>y0Go z45HsnJ;2`!gX(sRw3d))VeP-@$L~HsZ1v7Xg_hA(KatGVS)7(9j|eiWgNv{}M{6!# zTlD13&8+P9%*U-9w(3oKHKG=RW?|gY`fWt`9`sML`Bz`f+JRbgbdAR(ij%FpyFEB! zkf>t}Yr2$20}t838!51k@f2U_o&)kj3rke5!?+6eS7580&TmmT^V}ol-1zSOV#%=a z!E`R7Hjbx5RkI}r9+>WhR$?D}I?=V5AEuA8&{BFL`#Pf>D&uRyk^<0~r$8%uh+RNTr5DN4H){!0kz4vR42hGrqd}Q%wK`#&tCaQdEqPIk7 z5eEx0_>gwTx$4l9OnJ|&om3Jk|6d^c|9k&7$I@09YCnxYH}W0X69ZDo^h6Q@JG{UK z=tfhI6S+2oBgpxjWEX&vQ1o~+TyDL0QP8m&%_jc}R14otxS#`H@?5PNpdyV!e!{=G zaw+12T%jlSt1Zqt5a8GzM^(Cv^Hv-4OcZRBXWB!{tCMnZiAS}c1IL$g!(@(Z2OXL`7z5RD`4Q6;4xk|K@4fK!22S-q zT}iNn+b2Ejs+rH=NTnZA)x}$`au^aha26Us=23*QO$W%IWz%U!loAk1CiS!Y^Aua$ z{dz=jY1ipZ_```Cr!$cnZPU60h6p*$cisgnD}rnqz`o;&M&I)Nb>UY+Mjm{)UrRVl zKduGp%L@Rx|Er||j_!CI2riS|i#zi!H>Zpb$>S8`4YxE*aVn|V+l~wK?PkS87Zafo zbz%~p=_DTai-Q3a^HSuep8rjBPgTdasyIQtuSmt^Gm-x10vu8Del;IuaMSQl%0L^s zt=~$9^((6_*&i^u4}|3TyPU`n9%ua-%(W}JAqt=4R8W;NBd!Ts7#hNFL`TBAgnECu zIei0s-%{yo`hN6(zn~Ctd_#Nn6}A#UGV4pSjC8Bt^4+{cV1v&rFvwW4W*>jWQ%gg?bS3>N;c{~BxqCFhsP>`=cxfHXYAti&?JNHo4Me<$*u0tZ+OkmaGQ#1$2xrYjWhua;401qTzy%Z0 z|9wFhorRD`jZlMzlu^ty1$bUMjNapbKLfvl9Idf zgLRjWwTGcAJ_{K`eupO;L@7EH@dYt0gstH!x=H z&DdnUCH?vi_=+ix!W`aArV@YJUsz3k9Q6za3KEZ-Up!VI=)6NYcqQ@8Uy4T;@vIyx z?pmKgBQhjZ=cC2);~Qt#U%$?josD19>QwCwcGeQ?v~grPUt>=X<=qZUnp+ZpjlUYb z-g_70=93b+A>z!Cbx@E!L&C{}*@972_PBeqjx_Ur4)DFf($916WpZa8AZQlQHj|=l z{C(?G>7R-jVaJc9r>UEyDV@C#S}lKEPA!X#{!D<}&fduaFY*rY^yz|^H;)0q@%)B) zkLpu9+T&z&c#7MiF4U+-$jkH#V4WiRQ?*Z)uQZkov#g8oa@SpVrG8^@bF%-^*>GKvtdU2I5BWTzBk6ahRv3FrXRJexY^y0}?G& zt?6UybF83SH4)x|Rt=Mbf{Hh#o-!Nbn`XiO>X zLpe0L$7L#>5b{C&dA-y}4$J*2`*IzZdze(oD84K(EAwbC%XQ*bYzoSxy5Dl?y)1 zav-ZcmK|Q#+v-)THa3Qp4{g$bI%)1^t-WoBu4tX!Mo}mlqT^xdj4YuE8&M>pZk$=D z@C^Uk!!)Y<_ghO?wNO;M-&9fpC!Fu6GQAz##Fvm%ujc=JS*vM~7 zW_+X<2Nih=Iu^r5>EgyKixK{6p;MHulD$`9;U~z;Rn6X($K3PvuJyN zV4OO7suUc+QDZNJ@p?_jm5RX0?~p_4H4I&dW`_Y1aMO!)Hs zC>FTTe6)K|znLS;QwQAdZp+-!>hYoy*QBjrt8*6>gW6zp{sb;>CJ_2~r><(>D8`m8 zrWgOoe@JFutKLO;wJ%PG{r}3t0fE#{{&n{{SpK6|hD#a$pCAAL&6N@USfLO`x^I2k z23Iq#4m~mea(*#d0m?t;Y`RBZk^gQ60EqUd+hi2I45)Pn*ZkcQd(!DXp=lLd7z3zE zb^cR{dCA{HzGgODLm5W^NtePCI>1!&sb;*v` zXj6NiTy#(4Z6;Ux6AiFvAAPE8pa{`k#x5VHPIy|7GtUQ7r*+bKXv>LDq}$~%hui1t zz2}CwT@*gC&OiK+U<@b)X?HmL;IwQ1x1onej54dcOUu#GpPMhP&^kP918yu%c^L=Q z5)3cAa`Q&@sT+$^CNaRKNHQ}S+p0r0lA0u^>jEiXeyYu#Ji38SGqzr&G`OOtEVR|h zra>*auP-s^9^+Af<%pc|+a_xW5#9APUSs>}M60a4e#gU+!E(9Tl`QfH+frRV9j^^u~g=<9Kt z_=GiL#X)i2xi& zu<3c7JAj?H^E|!6n9o zO?DCjqT^BP0|LBg5#Kw?6RdSe`=n$WL@JYcRi0wClDGx0H)33-?55w-34fhWo`uyu z>xHqjfB7~2&R;_~)8?WBVE{Zb0EcqBB;UDOqE#nWf++FAn{LNgL_Gn2eqYc103U}3 z?niS#Oo+Z%B09^?7XrZD7rw{fF@UWIVWp72#1^|dExXWP-NLFHvnq}@<95>gTzI=8 zlFF|wp1vy!49&_K*y59ir!{OaBGz65$)Aysyx;hG9iLx`j{=IWplz zr`im%hy!>7ULFewBlD9j*0XBA?UaD;@^oMC;DfZ8 zB?FbHy<+mxkcnKLHgP#H-dn>aX@g!qSGth`v~^^|jr<%R!G*}zS18K3?>|+i;vtcw zdiY@WcUwfDmSGzFG9ZD+X%kQzn?-=;N-`MCl3pWo_nq+PFNJP zJl!=%T~u~c?OlXY_pYlMd2chsS}o9Gwng)^+7c-hOjewWVVzZIWRDa}iCku6z4Al( z7U@hs%+CXPvfOMNicQk{9qfvRsOWfQ1~~YU&$(Gr3hK?@vyrA7#c1(r4XEDz(EPeh zC*mc5wk+Z^kr0|#qbK8=G-o=z{&Verq-xZ}hD)80^Tn}%=C%own6EH313nKb$=jt= z*5nh$r}C|!qdIc zj;sI(o_4x~D|J(j@c{+A!)GjA90LV<-K0lZQ73;V_St8@8<5STn}0VDBcRTy?leOb zuYsY2$lAm>er$?wK@TPwALZZTGvJTv{J?Z1d`AgI#UCrsyx5n_7>8NjTFg1atAW^f z>tKd~hM|C02XnDlYxi(CEM(!#NDGE)?G}%Gx!Vo4Cmx77sFW`NYH(`%4aEd15FC+=V=hQG)RGl5P4cko zyqKZQgAdH00pqvza`Oflz}Z>@AGY@ebF$MF+*`#M&r2)pKT^MTOb)X^dao1q&Ki)B zvTedGyh}FG-+e=b%kUi|a(0UFfOneOs0l(CaT4g1-&Hu?m%Fv$_J&hO=L3%ND78!c zsGZMzdA5$LeAw@M#6GY4*J>$aoEuK3{tj>L_VF^Pu;jxD$a@A1 zs1gmTg=M2=z@b@Xv~`hd?jNdih3tBL3Uv=SoZQnc;$gM8c`&0YZ1f0xx(cPBdgT+I zJ?J%7eQ`X)LT3(`VFO5LWz5(7m;76ehbd31APo@rL$&+Vw;r3=jc4Eqia zB^k=ymihfQLH@5^e}&;B3e38O3lfNl|6-zuM-8Un?95) zg}^a9cqyXMlt{3Gtp1*KQJh`4+&YD6TYez3BuQL)oR+v3aKs3rF2iqg^|T5W`irH@ z`*$@YE;K%Bb2{!tKHQYk@cm5znz6F$f$S3}Ce_jN5P${d$8rYtHZPC<&n%cM<;`+%2_ zW1KzKv=W&5N<5>7lzLa2Y(@#u`E8dLzj^rAZ>D7Vn0}SPUo4?$N(rrA6)Zu1u2Y6oc$;lMXz$;{y+zI6R4hXKOF4EiH_$hT3&% zl=l0Kr&WidpTvB>npCsaTiw%8k2MTh3~{ODz(DGnFsYChMBHQ15sN za^}~PMA&(|(C14Lr}@|0#6f|l=%N;gLH`$NjsdO;T^wxxKF}W>rUEj^L2N6`DhGpn zJV!Ib!a|j2Fic9lSlS34PH_y@xG?5!U4%> z0Mb&L6vqWlg}my_9`U`h2#lf7=Sfj}D@LBe+20ey(qSZfjZj*1S>UI7^f7jV+vZjl z&l_bc`~YY6c}pc;4@9S#Nw4!YZk8GKGHgcS_DoTvek$H6EKoxm+l@<#aPqgIaw^6T zwfcsIvtsl!Lm)&A^8>{tOFo6)#7_OSjChQSq)IVxZw1j#bKw=K;svk^h(SN&{q>IZ z3o%z0zS|EdK6in?)>XqY?B>$q4ye$PRPXhzuqQsF1iVK9a6Oo7FxW$+6VL55zHyw| zBgT&@@%oN^`MV3?#H>j5@AUI@z^<*ST3rUF7g3?Wx}D+g0IoH4twYzzct7EN)_Uvp z_+V$!xPB+$@|^^uHatq_dV?z+`hBNlM<~?s`nj-07}IC(_KI>tL8V<1Q~{yam}P|i zZ4`@!Jmq1DfM^Nr$n}DJW*TPH*3P{gNVwEtWy?&y=8=$zD1L?%8Ru1lM>)+x_|pgE zc8E{Tc3EkRWIP0Pq~f!4-P*bh5+3wnKMT2UubhSH;K+}bZ8CYlCMWSt|PcgOns2c17o zIcEc}GE;Rh-^BQ7pbeM!15y5Ao3?(}md2r@U+KmrNvh_rY=lk1s@cun=zPg+z$RNq z);gjVlWHD)@EKW}z@^{TVj=A@TV}LntS4D7B?`IUz2q)30WRY zjxP2f+=Ym9QSPD==n^5+<&(Lr;Hz!4-Jw?Wqm_vQPQgheFQU{MrN~-``X;OLu?got zn4DIgr55BoJ|$A9T94r$+*wX^-gLEWcZa;AEb|M|y-e5q{1F{V&o9n>FCPQXybCd&~2XHtucb`6f8F({Yxin{b6}}4CpfAZ4=1EpIus`| zcH0-;Tn&I47jyI&Y|NG08KljhB>K3m(rdFvNV?B>bBg6HqdU2X4!>KWT0*S+F#M)n za-@l|=0PT5{c1nNv??zBy%$h~0C)ygKv3wsf$_`X4YvxZILR}W>HZDdYiuDjqc3VS z+f1q0zZyLQGs+SN@-`0xPlu-*$j%2%Io1%+YtiHN5)RRGc%P>PS!(zh3}pk6H&sN8 zG8lpl-EQZ^j4#R|tq$S}QYZ4HIiVELh$E~qyMu4yn4LXAD$pOFyeW4T0tn$p&_2v+ z5j3ODg}>XO3}S_}v(w$2(4C^1kZrcFZ7bjpH7Om|%a9|p0`V~xm5EdqG`aMomG_RU zmW$Ujp>kA`3NCMYVbCrMQurCFr|1@rVrW8Ee(*NRLMaPEMm>zDbw!abkn6FQuRot~ zuT&>d4^o>7qTU6ql5tV~f}~2Il%pEMENUN8-%)oc;bL)%SfH7DhcS*Sq(CpKe0PNP z2QDg9^4?W|gEWo3yj%f6wO4*6B&*w)|~y8nQN2|IC#W$gg?QW)!UuZU!z) zB2TBHa!^&}kynj;0LNb3A%nx_~7@Et-qq1IMMO@_$uD=H^@@w{$cl@_6*!y5inf61^VTG<;? zi~93({!BF!=S9825-e2nHETHV?@eDfVNWYa|4TN-?;(;m?TW137jutb`AjN7w zz!Xta6m551Iu_uypvxQGeB^{fEE7ZBY!1xyd)jR~PRnY8nG`xFdbyUUK^N zn?&$5ThB{Jmak@&*~yyRGA?2Txm3Z8cO+q6h}zNpaM-7Gtj3S%hARD^lpvEx+_T^U z?y8(~Obpa;j2={8)k1^_`7H^Z^)N?Le&)8pb#)LErP>hZ{e2t7s1*Kd20qlDCRB(xt9T>*Yoj(h z`D`N)w2E7Jwb=xdmzcbZGT)nT5gor~Wb^Ho0By8bR|0VT?gZ$cO4qzL==RA0P>x=> z1{U)}$-_Iq}CYM*d)1rD_t#v9R6(EaNp2&m()~~(wu0-4p7ozE44WzyF zmd8k5@>Z#iv^eHw>(4R|t3A9RwM&ls{i^g~dh7}S&3<8eM}nuDB0t*~b?=7WNm7lW z@3X`SkQSZMu;?>?6}1!2sazUDMSC0M=J`P47>A*;hSCD|R-xX7P0>+ch_N}@y+ zF|)M?moI0>Z{Yn^Y;><@owJEkj$c|;vF(lOZ{neopzem;FU_b$7~2kEShnY*v&00K zJV_cI%W5Lxmp{66hbUA?o93S@cpnbt-A_FWF7IN&%j!pa0<87~dn=OreODc}i`dAM z$ZK~f6Z0QHMhJSbl~P*#x&5ZjO-azL94h?}G6Wo`+$xe}`GK7gpuEdOQe^V|+WVb{ zt|ZroMJ|m7VHPH73%hoCJ(8Wwff^o0&7!QKaYca>L%6tmm@I9h59IBR>8&@U>m4J{ z{pOv|LxAI10}sVmtsbpndo46^V+(n9L^N$OQ+)e`NE#R@jP8ri;(rec)1ZTuNKpps zrp`n5F_>zw_Df4mGh)e6kJ(qd=L~TZ{(gCGgM9qVG13PgAWO+Bn_cq-4aE1ZheHNo zJ-r#rc3vDqjVTW|iBn^<@3f}q=%~7>=$Nz>gH4V{t7k*4Ld{-7lD*BJj5h=;2zumd zSAw@z!g7O}ZhQ)wq;D83y}M>Ea95_7ItVe{e!F%Fpav7n41xK%r~h&ABhYU!tl~J- zOQLEoxPX*GLp5?p+GHFTAuY#y+NOEoeX=*N2)_{J6ztF;*o=0nU;JjxJKAA-Vgn!S zu!F2bj_zUxT8msE&gx3V=;*`lyYBPtj2QE7WMHLy@px=rR=Fe-! z9Aq)Ous^@It(Scopxl6KTZVd+tSS;pk^)A`BZ1C|Do)e}%t7!V3HN)4oFP}%_~C#3 zp{6n;<8QFm?cBhwTQ#=+O1FI?=v($VYTv(Eb58V$l%)ktLeznOT_AsNPEVx)z!NjAd}a% za}L%f>x)?D@lWKL0&T_1JZJhk&ef1+9+%|9rYbaK4La#_%NuztLGp7u!5+OqbCLCa z9uWHHXw`|C{7X{FPIzmcID3NWJboHbKdr<*dt*q*Ae}ngxBl>o(~_Nup$2GAf8?dO z`GwCZM~+_vR?WPf?%0bS1xGJKA-%CUma;J6o*G24Y;4JTveWfdQ^OcEsuyW`YHIsN z!~~`LFJ*Gyqix>N%y9HOlah#vjt$d0XdDm)cLf>{DbSb*?6WmPj1-Nk&$HE6x~Au=eV+xZYCab-zvnUejpaxCSS_@O?onc0Vn4& z{VcfN>gN7Arj7!}^Jj+5;7We0Ts-#ZA7zY4lGIz?6ohBp(h(HQik$;1_g|U`C{ZRi z)?z8<9I~=nMQz-2Q9;QveQ^{#OZX#UVXU?>Rl>dJf7D(9;TSwWVXX<+9-EMxFPn=8%93i1T%T{Mt8Qo_`#l;e2(!B$uzrvjtz{*>tvY5{}SJ7%#5!q||qJ zwJQE?9^Y4Dl^2s*IQ41WHC{xI&A*q5+-5v*A3ad{K#i*sTu3H+fCjq5(k(u=Uf2j< z{#`%i99Pm7Q0Q!K3vPy%HXZwoNXA^fn)}h~Wb?V*%`BT4sRA>6+Sgdy{irQ}3rnie z_Llk2H@kod6wY1uNT7XSELi3)Wl(9k<6Tj+w9@w9#TEFt>usH+>|wRSC4| zek|a8znjkdoAHiLjFn3nI?>D3N$6Jzm4lNu4+qUtWo>L5k2)Vg{wRy~|ISgweYahw zhcBfIAIV_n9OO?s^H#b?dL^!D5jkM;m!Z}mwRwxIKNaAIv*PYwy-tNZJIELG?l0ZN z148R^=svWntCVzWLZ4A!^MOjXw>)4cK32zYvNq{o_lPVs;RZ|fJ2wZh5Pfo7QN^>y zFn2|7&KjCQjSorMeD853(>Mt+R~Zq&HmSbBAh4j>*WBW=##f6Q3Ms_v$s}2F)_^?% z6nXtBJ*%6Bn)@bcjF=zj49KB)VRZdNC`YGoj#6Jnouk`U2q?+`e?-6}|0MC;LR0lJ;d*2l zkyuw3PzG9!YSZBz-zJ(uLz?lYmK1S6c?#v^H}o4`q9qLIN9Zo{t1b?FZh-e@@zCgvny#ewh# zZI#hAHuQ&^(uFdeH{|Z9S+tkSnKaUg?&MJFBu zjsA(Y-^C}zKMz%VNQ=t;Qap;DU~zb8u0rx-dlSE*Y&hbHBTdJZkbz0eEPGH@%FI{P zuw+$I7%CQ|;d59pZ;G8}8%1-_s;P{IJT#MQ>cAYxB}cy+VC@p_sDcCatOxfyb$Jk! zj1}4|7?=f1;(Gl9KDm|_-3F{2WTcJsDn_pdccJ9jroB{M+;O-4uAtimjAjTz6f?6R zruxk6Lk>e8bJTu`yeM`U!BUZi!kSX_izU`?0xK}B$;Nuujk3IX^o*??O=%YvA|Wwn zgA%pQPwtJojz@Yz*ZrLU3ziRjnPvz;0%F!~B70BiS{i;ff3*FpVSp zP3FctabYh1cimhD+C}=t%z(s`TV}cbVC4;oQVFLRGo%MR@!rHFFC54UM_JLRJRWdwGY}!)LN$B) zwsAg1wFfGRyN=mX2p9YBLUw!2RUU4T8^n6Cyu=km8#-}**pvvssV{+VefCPQti2tM zZ4*?*TE*lOP-))myaC!|uB)M8dw>AKB=f_P5PK4U(6j{>V23bZEQ@opsTzcKh%Y`z zv`tm;h+q99#JJePSuy+ld=V<~62-5081;9V+nr-uD+%R3|QX{u!S6|Sm8Fn0- zm#=q09L;lLP*w(3+?e4w6mXO4`lVL4?I|o-ci;2g-Sc1OnO&U=3LrENT8x=GNO7=B zZ@*pgH3FMQoz#DpS`%PxOe$6|*@4WDXxddvFrhE85CKV!{@;$iWlW^7Hak1!viZjn ze)7c{mS^*O4}MF{=j;9$(u6-eEGs7-Ttc`u;;L})h(iqpA2?>vJZq!1jSk(|Co?l7 z)w&Cy6zphSB!w1lI23#}f_1U;vyZT)Y0>3fPQdsOb<=sv1;>tFm{blqA<{%?r6~U_)gfOG>SA^P`R{mldyxb1O3DqCytGmfRvH4yc+OUb0st=DtRu|x zw*i|0y48StpY27xR-JD`@yn%yM@f z27I5TM1}K-B195YI)l;V#N$#7_B=&$l&b9L@`SOwl!VGkyYI}Rstp=BD>W!`L5*-- z4RNxfv1*tvyJ63q-mk@7rh1E3CJ3lj@ez+tuFn6AXZ3QjW=5&56aGX0_(>|L zWz~17b942)*AyT2t0%(M86_Dv911Di$cLaZLH(zatp2Buq8ZCuVgEr1K=g>L7HFhE zfTwm!^yEM!?SrkBr=A|DV(B7yIn}AurVK;%f+s-J|7U42#4%}R-<#xH0AO}vJ+lY9 z3gmr+xf%$V8lWYSoUE$!N2cK6jw3r1Ek`bHjdN#T@d-CgbwcvHX`Y! zM4fufKD-2Ei8qy=;NZ#UPAAnbFe02O7Zz}yI$9!tYFYXinw*2+L;@wdD0cH4xKHMB zl7e#4DxyEims`_U*b$9bk)LPkomYeT^nOb|i5%91t8}^q=C-_SY-Y(XOB1~sa14t9 zcxYa?dT9%`t5A1QwO)bgsK*3gw%uWMG8e?~PE<7M)qpXVD}YbJ~*ocND1 z4RAU|Bq{{RYyQZ4kQ=_9Oo1C)L%dV%7XFLUC7=Dxz+@}U0b9($tWWv1a9qF7WN2;x zUSwsm%d|)IEUCd0W-N!`7U)vhMjM65a$OZAaA6okQMtbyxGO`A5_xy5?l{(%z6Ble z`$!D>Jk`BOSn#`qR18oI`QMQ4WLoE*-Y-gV9)v-4pOlK-KY=8&)$FIGL$?rShz45W zY+*FMtD93@tar`J9$J5t!Ud$`=jjE0wZTMUh3hNQp-7aKs1DEK`IDj~81g?E*zp1R zx}(R4;3*ha&DShp7>A57+^82(Q{@*n;X!8A2KIn)21qs+1hTBGQ z;10w&(D~)hwKXx3yROxcIVsGVYRiMr#PdKae+Tx4D>Sm%MetLf1QMtGS^cZ=1~a^V z9*znF{l;p8e6Ao%^Ibv=gtM9q@Q?cW5>Q%V5+>M-06DDjWVWFLN$G^DEy)mc4vnU= zo$X9&0NfxIHI@#0Qw3;@cyQM(!g~|=3j`_^<>5eyvD*fYiSn4aFY1!zH*dakM43`d z?Ylo3C13pKdlkCXUdbwmcX}M=C<+Ky z$U`}XOv#elmg(+W-=GCyxjX1Y)4S?9uHDJU`ckqVY(I0x9()O6RAuBilN+-YMMc5X zk3brg#1;lIFY(6}pC$Ls8Bl*T`;w5mmrf1Ke#}~T*8 zxV_Htj+I?Q3jHw{esQ5+RMcK}6x33!3|FsbLtv)l$;lmA<3|4&gJ z@Sh*w#LA>Q_3jGCPK!yoIm)ZZA+#hhddn9_I3jIzFo;g~5a zp~(mMK@-SB3si<|c&%}_GnjNZvmW*I*w4hbn6u7cnFdjRDVLTQ;(rf%D+6_m^ms78%!B{1taeqjEub|{{ zJ=q8C4O@9Pgp`p{ft<1*mZ2nl2%NF#S%{smWdJe}=R-_mc*Q|jTGOWrJ{r+2DN^$L zF)%5Z3yv^=+ZLGxC2|=cwL=oWC+JZ_mPGfXVA1dy@+9j0z)v3gWQ8)S?Jv;Dp_)c3 z@eqy0^^#F6@Csg+u`mkMMAnB(enpJ6LE}qQqU&g6Y(-5Psn|Azgmp&8LFL|_#(E^w z**-3NzRIZ9zZ-I{umG(6qRp&<%hF}m?Y^fMh*-3vb>%?ppv+wCpx6hqsyR3wQleW|HrlizxE_wzjFmH|MkwwjN%)0Gw*k&Rl zR)>hBx*}Vpp}nZHF+U7pW+_36q&9?Ey$*dHM~b?FAKnzusMV*F$PW!>e$~CsS2^t5P4GBAe?2{t@<>05*DHh?KaTDMBx+NG z%J&(E0JTmhhKK4WCq&x3tq1Bojxoa5Vg?kYBt-7oH1Dm+118VD;vgx3!;eX~m{pQq zuD#sIvBl7^n&x@Kj`+oXUiq3~I0zPYi(wnu>?*RQj!|^9EWUpxsZ!g#7_z7BKpqbS z>BY9|Yo9g{?irqfEO<<^)AOzRc1!5;(OWkyg@-`eB&TtYLLccsPwrmE1vPMv>@4z^ z={Q04{^G+(D`QIgG-$_skh6F%BA)Ow7(b6iI001$3VD|Gq2pG@{Iw%yJb(@L=IMEM ziiDe>63cd8^>L#&daqdwk5~Mr7`fYG>V98c*CNR z0F-&D>w`shY-@4Y$mZz&@dINifP17uvuz-n(xIvwOI)RpXv5 z$h>*!dFLw~b)4}SofhA>*)fS^_5|Y9lu~BA81g5j4jU-i^^+aMdyS)35~3dJR63&y zlOrBp<8TN|<4t-RnF~fMSyq5c;%WZ9iTByj5ST{3Q1pn0jlmPU zY3|o1?@oaj^Vs15V0FtKDM&hQ@_|8%W1*`nCy~qa+I#|KWrvu9WqrE4L{avy^S9&I`RB#^Mm2_T*W4^t(->=+`+8lm;Bu zL1ZuC9OzP+!u*0Q74h^*6$XCGPNv$Pd)HqzzN_IPXI-*gBnU3po>?Z6S1PcILY0_Q zxcr;0jW+5gNnt_l9>DH*vb^>#jT0Sx9LbLq8x= z_{XBvE9-GRSj|F*;1<-%kohAVV9p2Idxb;^KhlyuD|DY8vC3qb)xi!U2&hZN*(fxmE!dRaLl(w4DmqgQ1)1~vd<&gT*Z(P6wf5wS=C6i4S0ea?!R>ibo zB`GY97?Lol=~d}c!}A&EWH4EJLxb6Tk?E$EibS)kyx(W%-%p~;VH%l2xp|)n(td?O zL#yC5a3Ocu2ZKg_gJZg_aU;OTEF*#885YnU?d!R-(&D50SRuWTA!$4S1NOpKPDuD@ z@TY|mi2)xzg5&|!VD!|kt)&xgpxrWvBXPZZyk>DvcS5r%Ur z+m*FcRv`6Xw8>dW&Y~?Kdo`Fy&u$y^UuZ1ai3S=VEy2JP_TmqfjHSV+#H2;CQ6=8@ zK+EdUKLS1MUmA$d3*Sc~^7{8vSQCs;jvL6#)Uj{Ba=7p?j1y*`G7fYK08$Rnq#keK ziIDVX&}1VU*TI)$D)4V`a^+lER4eq`Sf`B%#hvm4kbe)QR`L26b(VNwP>g3SMWqJc z?TWp!2VQaK-y)DFkAWfi7|3=FTdVvQi~u!W(IY2Eg!njd zw?mQ40>CAMQr+AHUC1)M`tG5#s!aIP$L^z~qwmX0wVFiQ!4R6a0k!W!i8ou3AT|N+ z*MnXYp^G6@gMZP&+vzP%!b;RA85Xx6b1&K!%{skXaXYecGs5lwOc z)1@n`#*t;cljU8@reWT1^Na}w79#InTC6EK({okvwke_ZZa*WS=HS>_+EA;VrGer`fX+CHbfuZ)_i1MP2V`V-YkeN(&2=tpc?@-5c z$W0U~@Q__dK*VJ!SBX(Xx01AR4mbvgmxP43bXuF2QD+n^gi8B{UyECJ=_<+^*v}Q2 z{UJb?O(11}ow6`5M9PG|&BN|W+D<7;tzxE1e(k;WjCVU8RmAXXAe|)ShIGZYN!lFY zH5HQ{o2G);!S(SiI4uQ-%P(>Lo=Tn>5YSg&_I>2AIt;`aK!mpyD^<_SlWXH+&l=&s zmpeLiv3#_o8r%%FdLfTU(Ky-&z$~TML__d~YIgl;c)3495vhFR|4%AP08IM-ZTtkF zHb2<5vf}c5Y%7_rCaC?N0Ny1b+7zbeE75B>H*^`b;4slz3Fr1|SwywgB7E}(Ik&(7 zsQ_W$f{O%x*6GdbNG*b@J00002uIbGR44YvK9k4)e%8(s17_u5j z+;-~Z5v88Fo-!fi)d08e9c0H=4l8-K+q6FjsT=x;#jU)rzJyOCJ0bM4VS7b!C!=ah zp9~cXn$|cPE+0P6FHPeWo|y`vP%Zvwj5l-xzw|0%K(WKK#)#Hs9KL(xY*G-SggVI7 zCAo-djAu>No^;q7P@l{~F;90h07AP}?y@QnEIjCO^K{M>(0DrQpN?nzLDCA>0hq&G z_QySw1Aum{^#be?j{hoHLRzAoHuyomV7yczGuc;_Ai?a1`}tWcuc?w%*U(_%**^k`t2D!b=3}@7KDWD>jj=KzRJID|J#`BUp6q57y#}!O2%xZwur#Po-eFsYXC6u3Zkys+b7!| zqWv1LV@GMC&)Rj{%9aIA;IEYLM|v%KNGWwCCszfAjo5DEq>^2a9pQ!-=^ud_7I62^ zK1aiJk=o#Ro7c;x<%6}Xi^0R`eX>p0#bcA%w8%Oo{$^3PwfP$@93rU7ya-$T5R|48 zPI&+G3n*1;$`n_0uGjY`1jLV1RFTeg1}(tyN)?NWOuHMb5nISOCXe68BSeP)EVmnl z{ybfVo;Uy#Yw*6uflf@eU+}~k(IJ9rVX~nOLeYjk{UpxezqFuc=g2^XZ&!~2d62Jx z7Mu_HUD{6)(%Fhf>J=6qx>#wS>5{?79yp}6F#j_ba@`g|5kf1((3|#muPbX7PGBP! zGHdNoKrm%hI+RuM|H624Lj&+MoL(@5BKPIta+yz2J@SR#radt{c>&3xA zcaWOC#${&%l6yoRJ0Cj8+gNZwWfHx=N^4g_qEp~qS>@;xsMIuHvWe5!1uPi3YFYR) z9`bD=0qzZ&tnCKkGof&o5fQx}peH?J{Y~6+KdI%=;8QWb_}V$r!un*+XB9}I1L)v*M(mPiZm#4fL<$F+! zheW1#UN{5*4nkg;l`L}!Y5S%#sSd;tG5pcRJHrcb^Wk6#L?4{U z$}W*k5j|;?2D<;Eu=J=_bvZl$)|b^Wc%~QRH@MjSga!i7e$g4nrXhmAa!<;}gmiJm zAr5Va-yi9`WmbtJu0Foj0wk%-b0BOXe!>fQe!W_oV9Ku+Qiqio<@e*5KCtQQbuZ%I z3ezp*+NRZT74X5`G5~|pPF}{p-d#NxOZrR{otE;q{=Y(LxivSQ3X9vNIp|6Y$4$G| zb+7-wXijo5>%Gf3h8;mvQdjN+o+nQ^5*?oBrvHBBPfUx2QjIe>M3$C!W0U|kWqz*m zl4;L#;L)8AN zX!8ccDM<~*kg4piWbzkqAn4JG2k)6QpmMfDhyYCj7g?hdcE;ys)%b3bJ6sPFdiiwR zu#Q*AoOGm9D8K;!yMLQ1!`FP}9t55?H?PNVLtxmeUEiPjH)#Z5xODfV&4bn9FW=E3 z(!vWFu)aDYAQ{D!_9|UjAb~&xz{(gn`1V=e{wS--oT!fXmx}dQ>BWnXwoW5^r3ZlH zzxi)WTh%R>wndu#MIpPqx?XM~5O33cPQfI`%CKO~&>cIY%h5f-0_Npq;ljR{`(Qy~ zQRa&OlE#telCLHS!+s18&*ne$by&P>PNRr0?mQ`0>wEdFms)(RQOLp_w&$DaE@O0V zY5%l!@>s^a1{PXz7xkN6VlDpXslj0XLgzS~u8|M+j?1=5VIwq`@smon#6O0;AF;Nb z7p$cL+vRUI-|VIV995o$^)3SX-uAuedYAd=*dxaGg6;1VsAJ2hwo^^R=u?oHnN4}i zvq_PBkftgwB!{lM{`)6EM{6A?&!MOeFD!fzun;_iI5x37;|yIvY;i4H+y=K9BmYYl z9ZJ?|Fy6_93a-ZZy$UElH_)NEUKrGOtSN;*hpIsCpL;^>!Z~H+C?RfGl9EXVr2uHMkYxoz;n*w|e|1R+>_KH2kDp3TbU$ z{0W%$=Rj*;Jx}+FH@)=6uI&+AfBt8q#NW-J<$UoOaL>xPeDhTCsGlCvEVV`OLg^{D zRilj0;G+d^aw_()UKCZR(@#-Rx(U2TX`$69kdV`M0Hn5#Oxw+oA%AD>#U4tsa0Ux2 z*4>F_!!!AbC>adFjWxnyl9$LH?#BVkn{H?4&@s?+2lhbNoVBd-pMkw`^sS+I7O?U0 z>obQ{){F^~_h)sCLA`PfF32z^6Dl0Ff-G(WB5I`I#Es4{2d`6cn*UwENb#c}l%b>W zUy@MI9-AZxkQ1o@x!*O05lz#@NR|VudCRL#LAON0@~EyGpID8tK`x`xZGyP{Y3y%$#S-e}G9e#E(aBr61OjLfIZX;7)x;lp*j^b*v6j?y#yZ52230msD zDVgXO#Xz1&feZc*UXQ=yIWeu8}UG)(a%dfWWj_-z;0-Uj(34xt6=sX4Ew~r14}kB~{gLqUY$ozJR+f^~X!{ zO*4;Cj0BWta@618vxVs$Blm+JRkm1B| zXS@i}8rf0KA|yT6?F zVlsDvjDe*k=Hn|G-jvo8upfaq{FNNB_Z4@7s?FR>lSE!mRoU=7L<-d49V3!a*5Ui` zi}Rz-L5C*H5kXXG>Zc`(uHJGcnm?3w{mxqrB!Q0~mG7GGhYwt>i0E32O&`G>*$5lD zx|%VP>3XdIkxo?5+M|0#TRR4Gq|r+x1ZHRp$(p;>#@sN~j{k*Q^0-5yO6|b`zP=$; zwid-kz&WvqkH|VgQ)AQARhB2nyF~O1y%2mn8+gg;J}hBAj2m8@)tL!ohb5ePZ4C3W zi8o)HH|Qf(s*NI?Vrq%_F}RC5!uXUM1!bngh%3Qb8Z z=uo-Wu)m*Ap^=kC%r#RCd#yLZ1@CP^CMP?>rY#au*Ph}gjcI@E+8FBfLjlWI7RVY$!z9Zj#M@iVXA>PTl5t9>v%XOIw*as#*G4OX1D z1;?uqe~j7S()!Crr7Eql$|<%PavF31E<8xwDHZT=X-7*%i;`)lUz)3P1~}MJx>%U@ zUC3#nW%bJR{`oCCL9AQ9+9VZK9lA;r8P(IiKXQS6P2R-@1kpft1fGw)N)uBOo!l&n zV6|mbt&`SeD-=36BNHkOlKJ}bMxzkxItPq$-WmDbdjT4L`g=Wyha!w0C4gG@>!RPo zqIPK1;($T_q3JFSf?J3|LV78T$TlE!W?TWES^leMjYZ ze*%+#x}%3Ks1hs$iqQ94nXlHaWA>=R1g|YDi$6uHdMYIM^3J zg|sYLQOkOD(wVvfVCfuTCt}p>(Hjv6`?IiZ>G-v7@pqUAYKWBGe$B^PfXpGd>*kj< ze&e2FdtBlk!Xt=20e{p%Ljx-ZENS^yMHUgrn=o9LIj(JcOk~ILz09|Ii-7GV5t2CX z^&M>Ja;l=WC9T57aSzyB3_$b7s@Vw**5R5AnRR8S^b8Mpws?_C0ui6oV?w7?a{_~}vG@syBD0(dV+bt`#G^1SRFh97OeSfMjQ7ifJ zcb9p)=*Qmd2qS}-V@e1@ss2XDa-ohI_)I0_RqYeKg~_-$@*c3dIw4Q>DTvWJ zov%#UyTG9-6D{zZx?!Nh2w*kdgs!4tmF;c#U(}+-7F^y;Jk;PNSdpaJ+tr(yx&8WH z$eq$x5=V-Ftv|`C>T_uq^)iHR{gm^t*gWfbrFy>_xvQP3Vp--0CRD?OJR8F}_qKV(j+LY`|AoHBAr=#g z`>ca(E{>;-M)KL0u&BhM4tzbff>|C75Nqxj`a}TX7DZUw&+UTV)b>!#5F7HO2Q$|N zurE`-CWk44FywScOiml^N$&3N8|e-zX;ckB;&&6m2>fVhg=`pAikoTTwJPOf z!D-5-Hhy217q*O0oOo|j{fI5I%T7j)5AIn%LvA*(mrhnxP{`>HsKSKjA!z}UgEo04 zFJ;HLj%8O{DCG#Lo2i{}NHVS#QDRt2lb~>J^zAFVaEu>{YW*op0Z_!M6T_tkwGNdgn3IqJ@PkhIE$;;xjM68BwFvtFn4Fo7&Xl2 z62!65oqEVDg`4n)FN%Drk}HBWeXoSGXF!I-4cZJZy}WQTg|$S+)mf#nQWv{XjyR*B z{#XB^dIhG<;3&FVL_>6azb9ZEvgd16a-$L#$BmkeD3|3X1dhMa6?n&2En-}nfdogU ze}H-{2^HS_@~`;!oTg#s38ur{P|Ze+PfqS^8T7Z?ZlLTD*11R@L9UE<`@9~b+O?WW z5|4Q`fgAzMMien-ys>L>Nftv&^ng}9#nJtL5uHx&vF7(882dHd}AZ}G?@5$|OK2oC=Kgln25XnQ6RatvtuL8LQ zAaZHfvM_nD)-WrXI>VS2!Z=ybF6#=bU8)W12N^n;zq)gf;jIv9?aYq7XpLeAj7Imd z*$gZ@b^P5g=iYLi=h2$R8dvaJj&SFLqQ$wt7%B8IWlDQbO01}vBk#2yT9&Y)BObno z1+9@WOx^!ExhrX-JLN#ucSEtA(-L}M(8b8~D>xE=b>WY&kB}9y8^gkwxI7F*z}kjH zaZ6^$w5D>ca=I})t^t`thaFc zpX(9Kf|?99i`#)IPhzNkophn!vXSSb@YT-k{Q@VzUqi2A|5n!$ag2W3c4K+|)Q{gI z-xOf46%^P_dd9oM5RcJAd1PhcuE7Hl{CB}$lR?Oo;*5~aR3IjO+aY_lM*6^k ziI0D$or*{0(=MVkgCUGH&eyqs#*>F>w+OPP0RgDhbY$g>J-#W8=ZRmBJ5b+l2j zEmzhB|NFlUqEB8Qcg!5HrV+MU2``qtrPNU+3bI&dSz}xfy?5KzQzJiMl5c+dxnNe; znX7Z6A!u+|a~h>qyO1FFdOej}o^*WM5yqm-LtnV2_LAy2XW$zdRC8^+>K1K7l25ayHmJ0 zZPKn`oxS7}L+^?sm*&t-CC`GJ^KVcT!9v{lWtGtA66}WfiI#pNttXiDBX=`qf7GY; zeY-ImRs4wRJh~~!zVL6nmXUYXMoIvqO_QHCr@%=s)nK01w!nRT60TvLz0u7Trl&@1 z{tPNhjOe$yMyb1~C4w)EFhX8N{WUX4`=eWfzX5>TT1$BY8T3ly!bxp#2}K+K{5e`2 zU=uF3B2Uzi{$q)hyo3x+*!nIQTO8_#7>n$r)FG3AYa)&UYv~qsbNdnrTKs63CnFtb z!8zNeK8#6O*k_9Pf1T^B6OIWa9gFsPU(fFwtR*LXn?xuzeswW`P|8)3k1b_!VtHv1 z;#zmpuQz&sf1e^40`Kqn0Qx8lYe9Rab~Jn)N`{WYxYY|TWk~!ed%_s}*8$6I!lOzs zii04rw>3+@-yxz6Cdo_Hx7{;{WI{Z#q_6~I)zBMcAn=Og*{n3oKLde%jho(t( z1C?eTj6yv6$Zye)lzQp|#ps#|+HC?P3a460^0Sm5l@nf|!M^f|7xF-l{fQD% zmf*4d4v7RHus23BqrzYSeVP@#=t z+2IEzz+3{h824(RpJr0%N(VlKTmq&}!9c7=!NJ@^LhakwCU8czYAC_11E?SUAK=f@ zI%m+?Ell@eC``e*o6ql`bO^7frdQow01~7NFwa_O@dNc%1*l^kD{4-EZNLNW%dv-r zTInS4|F_8A-LlNavNN1kgR~?-T^21DeRe)%(9!7Mv{fo;{XN!5Y5@e2Lg<5Z-uuVa z#$IJ@WtynTXrI3yt~-Q3l!J-GW?m(;{A7FDwm9Ym2E1entLyAgr2lM+=?N3{?wn)7 zZ6CZAAAdPGP#Y??2{x3QxVzl9i)U2W$}0!2l)Mi<2HWN*2O7gQ4jEH5#H+*RlgZ;K{c9Ks2TgihTD`|R%B zuvKKC>QBEZm(Y-v`AwFrCu@X7)EkO ze6R_#_jC-(e|$Zc+u91(GuyDWt-u9LH%-~*108plb}`u-z-aQGy+wBY@7iK#!;smw z$|=*GMi%WB7ruuC;hun-UxIU9$zt1+1(w3J#F1D=+x|Jb3Gwd#fO?y&S=+^u-ST3z zO~JI1HA(2c*uA4pVNmjFF3dme+zhsE-Zq+*xQDjkw9mBHFa=LOoIFMRS_t+!Cd;l$K%ViJca0qrAs}IEGmlvFy z1CV=9{H~gsSs*VXCw^ZA6yNCC2|a8wGoQ!wX=T+-D6AfmiMaHjqi{s~DV0x6cihya zBJAV&GX_3(g)c%R1CzDzV=FjJSNiOw{|1>IW!Iz6I-uJ250MOUjU!InE&YY2AL#dK z$ZhGCsagun4{U?r%wAn>A_!MHGpl@1)cKAva)&xkXIx|nNMG+`o{W`wDLg%XA+?+c zYma1m;sKNBE65>qe60{q`+>MDxA`TjOMoS$SiInwOm7dSVkv(e9MoZ8ZqEK_iQp&S z0+_REmCIPsnvM&LYelpy%fn%*)N$@AKYL;SQa1f){w$m(lqb5aNaKLZvk?p5vpOb* z-fU0%>SGz!b^(sq4710wY<$mEhRyW@yJVF`?u?p7?eg%$)^s$zj=A*AAp5Vf|6*v9 zZ))!XP;8?v*^C8tFMvtDwiSx4H&a_mp3yZ%9%_d}J48s;==5BpOALlVVpPL;POip- zWVl-6ssj%K?glHXhbP5}FANj<&z^80WnKhUf=uJm&Kn}+JPZPmA;O$|L}Qo4Yzh@I z2nfEMfH%W0t(4E%X(L_a4$IS~ps5|Ub^}q<{tKqY=iP`b1Pxu)zn|HGJ<$+s|YlI9?E zIO3Lqgv7MdahE-2X}!aZ?Mtr*Xc)qVJY!fWCM_{2(CZ435&Ky?{US_Bv@?VN6V&;PTGx}Iv|v0 zXeh7?zTAhio-()VIaR3Bm}^Y;e74pUDU`hPGBiUjQvYTRxU$T4u?i6eEQOt|loI|{ zV{^H%x5_d%Rw&bs$*j3oRV(eNr+=%Hi%d@F-{3Y=?7kNzT)l?f);|1rNsVbuwjZ<) z0h#a9nFJiA0-+W5ag3e*mbCcVl@Rk81G6GyAM4mZA06jEEJzB|_Es}FDL32)IdokU?r6uZ>JC(zn zL2VTH=-(xeZNywOv~2(XYMQswN9C=1xcv(aV-WL6+eR?~yHCok6sdqLHmoKuXtBII zU=Q}to&+RjQ}}!AspsT(qou#mxUokW*g+;d@nZv2f)(VTCkoZ8u+8IS0000000000 z00000001$1(Te2S*p8b8l_1a!zP2OxU4HV~S2#ZR5>I(MW!1@^Q*A}gH+M4!y`I`3 zg+hSbA~!B`U^4X;1_XymndFxO(8KD#F6wD7=_U60%o2Z_%nK9)Mz9TTn$1V6(5-Yt zYVsPi8L^{v;KC5_=gYT$XMm;EPE4{=L+P-o;_#gN5&vjZp&{r zL(Y_HHo>OzYnD5<|Cc@yG&Uu(-N=RE{L+pL-q%oSw2_FDT|P=Txnd#}q}5tDQq1_*oyp@y*0 z2KGMTR>ROOyByt=@`wxoAW7bc$Uh%nNh8;@J~%!HG=uVvsTilAN{g4R>{DLCDbi}A zEHu}5D(qu=?GSM`%P}t}-Z_>sD#T1TA04T}WsaJ>C_4WML=6I5ToF(-C4Jbxrg;XJ zPG^o5Hww!0TXQGaroG96t#YSY(JY$aG+QUe7bUTG?eq|Iu&@ z@GiHwtYbqLIPSNwjgB7 z)tapk#mr$|Q2S$jj#JAx;T7X~l2HbwKfKN-5omyLcROb{$TzI#u}#WGixwus(%>h` zK{VBL5E-pS0x|h#PL9iDSqzmU0&v*;X40bsWh?jX{eKo@CxQhL&u4rOu_Xtaphvw0 z>ecYKDyg;ZS>E?sbpetUeA+KBt^Ch9#j#6JDe47x89`0KW#JE=cC`P7UHaL>_nkH~uRIQ!6B!o(LgMGM6GFDM5z+Y4 z6xGr5rFc;>!jMNnIXycjwcG^1h$yZ2Pf=R&NJz20S z`OjuL1bSv0{<0%WywFep00000HH6h-S%M+kCRLMMfZvrMI%YA1?f|HsNP|#&pN8^^ z0>10EG*c8`pE7g4p#T5?06qbvo9r|NPO8CJ@)k7xgmDDFtxZK(bg;LVG zgcn6m2rGMv^eE4G2&v%(EVxLnxp1+_*S2^;TijP%t-^-p&{QgxmDDfglzU>V??J?e ze6_)wY*n4vsc9bz&!XlEDD^_6CaWCX+o^p+y#+@5&NMFAt2@P38TUVNeX&+|W~HQj zFFuQyD5KR1l$xw_cW$Ni4)hfp?>NxAVyy2KSZCb*#rDNn-I|t>@VxpiV4{yyDpG2( z&E2|})H~d(;q%TkF4(I(#%V4}g3gkFdq^aC1>LC%cidL@PDu|u#wGLk30CZK2~)=V z+n*)X>w*8XwIY`wre!RHffvTje6-MfVAikU_0`CULX?mtTd~L`PaE%Ue3w_|h01+p_BJ(B;1QP)Ee;rt^t%N|@yutQbL2Z%uff(JWm4EB9J;{Iw zli-nglLi6_fO|iVtXEdTAZ=b?`LII8gZv!{lo5vq;s@%ab-t#(?!kcee;bsoKDOmX zs$E~2A>;*B(hdn7USRvDEoCuj^lHP&X-(}=ElRtEzV~3jdq0iJR-apPqg5`i%@Fbe zt0ZpMzmx~F`1FlYQ8PUbad5}#i!A$RP53xB&hrb1QO9SWVkZBg0hRzm}y3CT@j7S%j*3pJGwzd2pi68 zm26V6(^oq5zgixT+?JGSeeMVYGz{!o+o{l#?3(t1&lfNX!diQD6O+#4GM-=`O^TKW z3_<5!?;=PWo1ukK*tZ(h4nLT$&x?+vr7n8e96p25RgGYQ?}FS^usC85IkZfmB>5~a z(D%wFh-cIxV$wf;GN_#(3Dpiin5no6VLtZ^EI}$KCuLop?cSG&b2KI`4XWgi;x)&n u#Y+Q*AoH6Dzpn|L7wCKC72u+xR%H_>cweE5^1TR&<&%Varva}x00024JY4Po literal 0 HcmV?d00001 diff --git a/versioned_docs/version-4.0.0/running-keploy/ui-capture.md b/versioned_docs/version-4.0.0/running-keploy/ui-capture.md index cd7668117..decd8c72b 100644 --- a/versioned_docs/version-4.0.0/running-keploy/ui-capture.md +++ b/versioned_docs/version-4.0.0/running-keploy/ui-capture.md @@ -71,103 +71,20 @@ Open **UI Testing → UI Flows**, select your app in the header, and follow the ### With the REST API {#set-up-ui-capture-with-the-rest-api} -Every step is available on the [Keploy Public API](./public-api.md) at `https://api.keploy.io/client/v1`. You need an API key: `write` scope to create or revoke an ingest key, `read` scope for everything else. Replace `` with your app's ID. - -#### Create an ingest key {#create-an-ingest-key} +Every step is available on the [Keploy Public API](./public-api.md) at `https://api.keploy.io/client/v1`. You need an API key: `write` scope to create or revoke an ingest key, `read` scope to read capture status, flows and journeys. One call mints an ingest key and returns the SDK's endpoint, its install command and an `init` snippet with the key in it: ```bash -curl -X POST https://api.keploy.io/client/v1/apps//ui/ingest-keys \ +curl -X POST "https://api.keploy.io/client/v1/apps//ui/ingest-keys" \ -H "Authorization: Bearer kep_YOUR_API_KEY" \ -H "Content-Type: application/json" \ -d '{"name": "storefront prod"}' ``` -`name` is optional, and so is `ttlDays` (1 to 365; without it the key does not expire). The response carries the key as `apiKey`, the `endpoint` the SDK posts to, the SDK's install command and an `init` snippet with the key in it, and the `next` steps. `apiKey` appears only in this response: store it before you continue. Creating a key needs `testgen:run` on the app. - -#### Check that capture is working {#check-that-capture-is-working} - -```bash -curl https://api.keploy.io/client/v1/apps//ui/capture-status \ - -H "Authorization: Bearer kep_YOUR_API_KEY" -``` - -The response names a `state`, explains it, and lists what to do `next`. It also counts the app's sessions by status and its flows, and lists its most recent sessions. - -| `state` | What it means | What to do | -| ------------- | -------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------- | -| `NO_SESSIONS` | No recording has reached this app. | Check that the page calls `init` with an ingest key for this app and the right endpoint. | -| `RECORDING` | Sessions are arriving. Each becomes flows `idleAfterSeconds` after its last activity. | Wait, keep using the app, or end a visit at once with `POST /ui/v1/sessions//complete` and the ingest key. | -| `PROCESSING` | Finished visits are being turned into flows. | Check again in a minute. | -| `FLOWS_READY` | At least one flow exists. | [Read the flows](#read-the-flows). | -| `NO_FLOWS` | Visits were processed but none had a user action to make a flow from (a page opened and left untouched). | Record a visit that clicks, types or navigates. | -| `ALL_FAILED` | Every finished visit failed to process. | Record a fresh visit; contact Keploy support if it fails again. | - -A visit that can't be processed is retried with backoff for about two hours and then counted as failed, so no state lasts forever. - -#### Read the flows {#read-the-flows} - -```bash -# An app's flows -curl https://api.keploy.io/client/v1/apps//ui/flows \ - -H "Authorization: Bearer kep_YOUR_API_KEY" - -# One flow's steps and network calls -curl https://api.keploy.io/client/v1/apps//ui/flows/ \ - -H "Authorization: Bearer kep_YOUR_API_KEY" - -# The journey plan: how sessions move between pages, and each journey's share of sessions -curl https://api.keploy.io/client/v1/apps//ui/journeys \ - -H "Authorization: Bearer kep_YOUR_API_KEY" -``` - -The [walkthrough](../ui-capture/walkthrough-dashboard.md#read-the-same-data-from-a-script) shows what each one returns. - -#### Manage ingest keys {#manage-ingest-keys} - -```bash -# List an app's ingest keys (never returns a key's secret) -curl https://api.keploy.io/client/v1/apps//ui/ingest-keys \ - -H "Authorization: Bearer kep_YOUR_API_KEY" - -# Revoke one; pages using it are refused within about 30 seconds -curl -X DELETE https://api.keploy.io/client/v1/apps//ui/ingest-keys/ \ - -H "Authorization: Bearer kep_YOUR_API_KEY" -``` +The [REST API reference](../ui-capture/rest-api.md) covers each call, from minting and revoking ingest keys to capture status, flows, journeys and test runs, with the answers it returns. ### From an AI agent (MCP) {#set-up-ui-capture-from-an-ai-agent-mcp} -Every UI capture endpoint is also a tool on the Keploy MCP server at `https://api.keploy.io/client/v1/mcp`. Connect your agent as described in [MCP client configuration](./agent-test-generation.md#mcp-client-configuration), with an API key that has `read` and `write` scope. - -| Tool | What it does | -| --------------------------------------- | ----------------------------------------------------------------------------------------------------- | -| `createApp` | Creates an app. An app for UI capture needs only a name. | -| `uiCaptureSetup` | Creates an ingest key and returns the endpoint, the SDK install command and the `init` snippet. | -| `uiCaptureStatus` | Reports whether recordings arrive and have become flows, with the next step to take. | -| `listUIFlows`, `getUIFlow` | List an app's flows; return one flow's steps and network calls. | -| `uiFlowGraph` | How visits move between pages. | -| `uiJourneys`, `uiJourneyImpact` | The journey plan; the share of sessions a set of its journeys stands for. | -| `listUITestRuns`, `getUITestRun` | Read UI test runs. Today a run is made only by a flow page's **Generate Test**, and it stays pending. | -| `listUIIngestKeys`, `revokeUIIngestKey` | List an app's ingest keys; revoke one. | - -The server runs in tool-search mode, so these tools don't appear in the initial tool list. The agent finds them with `search_tools("ui capture")` or calls them by name through `invoke_tool`. - -Then ask your agent, for example: - -> "Use the Keploy MCP tools to set up UI capture for my app `storefront-web`: create an ingest key, add the capture SDK to this frontend, and tell me when the first flows are ready." - -The agent calls `uiCaptureSetup`, edits your frontend with the returned snippet, and polls `uiCaptureStatus` until it reports `FLOWS_READY`. - -#### What an agent sees {#what-an-agent-sees} - -Flows returned by the REST API and MCP are meant for an agent's context, so they return the shape of the journey and leave out what is most likely to carry your users' data: - -- A step says whether the user entered a value (`hasValue`), never the value itself. -- Request and response headers and bodies are not returned. -- URL query strings, fragments and credentials (`user:password@`) are removed. -- A path segment that looks like a secret or a personal number is replaced with `REDACTED`: an email address, six or more digits (a one-time code, an account or card number, and also a date written as digits, such as `20240115`), twelve or more letters and digits in one run (an invite or reset token), or a long token that is not a UUID. Long IDs that are not UUIDs, such as 24-character MongoDB ObjectIds, are replaced too. -- A flow's name and description are scrubbed the same way, and an `aria-label` value in a selector is replaced with `REDACTED`, because it is page text. - -What is kept as recorded: the page's paths otherwise, and the `data-testid` values, IDs and class names your developers wrote. These checks go by shape, so they can't recognise every piece of personal data: if your app puts user data into paths or into those attributes, it reaches the agent. +Every UI capture endpoint is also a tool on the Keploy MCP server at `https://api.keploy.io/client/v1/mcp`, so an AI agent such as Claude Code, Cursor or Codex can set up capture in your frontend and tell you when the first flows are ready. The [AI agent quickstart](../ui-capture/quickstart-agent.md) connects your agent and walks through what it does, and [MCP tools](../ui-capture/mcp-tools.md) lists the tools. ## Install the SDK {#install-the-sdk} @@ -228,7 +145,7 @@ Masking happens in the browser, before anything is sent, so Keploy's server neve - **A body the SDK could not redact** is never recorded in the clear: it is recorded as not captured. [`KUI_E_BODY_WORKER_UNAVAILABLE`](../ui-capture/errors.md#kui-e-body-worker-unavailable) describes both markers. - **Visitors who send Do Not Track or Global Privacy Control** are not recorded, by default. -The dashboard shows a flow's recorded URLs and calls to people on your team who can view the app. The [REST API and MCP](#what-an-agent-sees) return less. +The dashboard shows a flow's recorded URLs and calls to people on your team who can view the app. The [REST API and MCP](../ui-capture/agents.md#what-an-agent-sees) return less. [Privacy and masking](../ui-capture/privacy.md) covers every rule, and what each mode still records. @@ -257,7 +174,19 @@ Check the browser's network tab for requests to `/ui/v1/ingest`. If there are no An ingest key records into the app it was created for. Create a key for this app and use it as `apiKey`. **Sessions arrive but no flows appear.** -A visit becomes flows 30 minutes after its last activity, or at once when it is completed with `POST /ui/v1/sessions//complete`. `capture-status` shows `RECORDING` until then, and `NO_FLOWS` if the visit had no clicks, typing or navigation. +A visit becomes flows 30 minutes after its last activity, or at once when it is completed with `POST /ui/v1/sessions//complete`. `capture-status` shows `RECORDING` until then. A page opened and left untouched still makes a one-step flow, such as `Navigate on /about`. + +## Where did a section go? {#where-did-a-section-go} + +Parts of this page moved to pages of their own: + +| Section | Now in | +| ----------------------------- | --------------------------------------------------------------------------------- | +| Create an ingest key | [REST API: Create an ingest key](../ui-capture/rest-api.md#create-an-ingest-key) | +| Check that capture is working | [REST API: Capture status](../ui-capture/rest-api.md#capture-status) | +| Read the flows | [REST API: List flows](../ui-capture/rest-api.md#list-flows) | +| Manage ingest keys | [REST API: List ingest keys](../ui-capture/rest-api.md#list-ingest-keys) | +| What an agent sees | [Operate UI capture from an AI agent](../ui-capture/agents.md#what-an-agent-sees) | ## Related @@ -265,4 +194,4 @@ A visit becomes flows 30 minutes after its last activity, or at once when it is - [From first visit to journeys](../ui-capture/walkthrough-dashboard.md) - [UI Capture errors and warnings](../ui-capture/errors.md) - [@keploy/capture-sdk 3.1.0 release notes](../ui-capture/capture-sdk-3-1-0-release-notes.md) -- [Keploy Public API](./public-api.md) +- [AI agent quickstart](../ui-capture/quickstart-agent.md) diff --git a/versioned_docs/version-4.0.0/ui-capture/agents.md b/versioned_docs/version-4.0.0/ui-capture/agents.md new file mode 100644 index 000000000..26fcda1ae --- /dev/null +++ b/versioned_docs/version-4.0.0/ui-capture/agents.md @@ -0,0 +1,127 @@ +--- +id: agents +title: "Operate UI capture from an AI agent" +sidebar_label: Operate from an AI agent +description: Credentials, tools, waits and errors for an AI agent that sets up Keploy UI capture and reads flows and journeys through MCP or the REST API. +tags: + - UI testing + - UI capture + - AI agent + - MCP +keywords: + - AI agent + - Keploy MCP server + - UI capture + - uiCaptureStatus + - service account + - personal access token +--- + +import ProductTier from '@site/src/components/ProductTier'; + + + +An AI agent can set up UI capture and read what it records, through the Keploy MCP server or the REST API: create an app, mint an ingest key, add the capture SDK to your frontend, check that visits arrive, and read flows, journeys and test runs. This page is the reference for building that into an agent's workflow. To try it first, follow the [AI agent quickstart](./quickstart-agent.md). + +:::note Beta +UI Testing is in beta. Recording visits and reading them as flows and journeys works today. Replaying journeys as Playwright tests is in beta too, and its runner is not in a release build yet, so an agent cannot run tests: see [What agents cannot do yet](#what-agents-cannot-do-yet). +::: + +## Credentials {#credentials} + +A person approves every credential an agent uses. An agent never needs a password. + +| Credential | How you get it | What it can do | +| ----------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | +| MCP OAuth sign-in | Add the MCP server to your agent with no key. It is answered `401` with the server's OAuth challenge, and the agent opens a sign-in you approve. | Through the MCP server only. Scopes `read` and `write`: every UI capture tool but `deleteApp`. Revoke it under **Settings → Authorized apps**. | +| Personal access token | **Settings → API Keys → Generate new token** in the dashboard, with the **Read** and **Write** scopes, and **Admin**, which only organization admins can give, to delete apps. | REST and MCP, up to its scopes and your role. | +| `keploy login`, then `keploy mcp-install` | The [keploy CLI](./cli.md) mints a personal access token and writes it into your agent's MCP configuration. | The scopes your role allows. It expires after 90 days, and is reused until then; minting a new one, with `--rotate` or on another machine, revokes the earlier ones, and agents holding them lose access. | +| `KEPLOY_API_KEY` | An API key in the environment of the `keploy` binary. | Signs the binary in where no person can approve a sign-in, such as CI. The binary has no UI capture commands. | +| Service-account key bound to `UI_READER` | An admin creates the service account in the dashboard and binds it to the `UI_READER` role on one app. | Reads that app's capture status, flows, flow graph, journeys, impact and test runs. It cannot create or delete apps, list ingest keys, or mint or revoke them. On the Free plan a service account can hold only `MOCK_READER`, so this needs a paid plan. | + +Whatever the credential, what it reaches on an app is what its owner holds there: minting or revoking an ingest key needs `testgen:run` on the app, and every read needs `app:view`. + +The **ingest key** the agent mints is not one of these. It can only post recordings into its one app, so it is safe to ship in your page's JavaScript, and the REST API and the MCP server refuse it. + +## Connect {#connect} + +The MCP server is at `https://api.keploy.io/client/v1/mcp`. It runs in tool-search mode: `tools/list` shows only `search_tools`, `get_tool_schema` and `invoke_tool`, and four sign-in and setup tools. An agent fetches the schemas of the tools it needs, by the names in the server's instructions, in one `get_tool_schema` call, and runs them with `invoke_tool`. [MCP tools](./mcp-tools.md#the-server) has the details, and [MCP client configuration](../running-keploy/agent-test-generation.md#mcp-client-configuration) shows each client's configuration file. + +An agent without MCP calls the same operations on the [REST API](./rest-api.md). + +## The lifecycle {#the-lifecycle} + +In the REST routes, `…` stands for `/client/v1/apps/{appId}`. + +| Step | MCP tool | REST route | What comes back | +| ------------------- | -------------------------------------- | ----------------------------------------------------------------- | ------------------------------------------------------------------------------------------------ | +| Create the app | `createApp` | `POST /client/v1/apps` | The app and its `id`. | +| Mint an ingest key | `uiCaptureSetup` | `POST …/ui/ingest-keys` | The key once, the ingest endpoint, `npm install @keploy/capture-sdk`, an `init` snippet, `next`. | +| Add the SDK | none: the agent edits your frontend | | | +| Check capture | `uiCaptureStatus` | `GET …/ui/capture-status` | A `state`, its `explanation`, what to do `next`, counts and the ten most recent sessions. | +| End a visit at once | none: an HTTP call with the ingest key | `POST /ui/v1/sessions/{sdkSessionId}/complete` on the ingest host | `{ok, sessionId, status}`. | +| Read the flows | `listUIFlows`, `getUIFlow` | `GET …/ui/flows`, `GET …/ui/flows/{flowId}` | Each flow's name, session and counts; one flow's steps, calls and end state. | +| See how visits move | `uiFlowGraph` | `GET …/ui/flow-graph` | Pages and the moves between them. | +| Read the journeys | `uiJourneys` | `GET …/ui/journeys` | The journey plan, each journey's share of sessions, and its `freshness`. | +| Share of sessions | `uiJourneyImpact` | `GET …/ui/journeys/impact` | The share of sessions that took at least one of the journeys given. | +| Read test runs | `listUITestRuns`, `getUITestRun` | `GET …/ui/test-runs`, `GET …/ui/test-runs/{runId}` | The app's UI test runs and each journey's verdict. | +| Clean up | `revokeUIIngestKey`, `deleteApp` | `DELETE …/ui/ingest-keys/{keyId}`, `DELETE …` | `{id, revoked: true}`; `{deleted: true}`. | + +`uiCaptureSetup` mints a new key on every call, because a key's secret is shown only once. An agent stores it at once, and lists and revokes old keys with `listUIIngestKeys` and `revokeUIIngestKey`. + +The snippet `uiCaptureSetup` returns sets only `endpoint` and `apiKey`, so the SDK runs with its default `privacyMode: "balanced"`, which masks every typed value. For a public site, the agent adds `privacyMode: "strict"`, which also masks all visible text. It puts the `init` call in a module the app imports before any data fetching: requests the page makes before `init` has loaded rrweb are not recorded. + +## Waiting and polling {#waiting-and-polling} + +- **A visit becomes flows** `idleAfterSeconds` (1800, 30 minutes) after its last activity. To end a visit at once, as a test or a CI recording should, close its tab and `POST /ui/v1/sessions//complete` with the ingest key in `X-API-Key`. `` is the browser's own session id: `uiCaptureStatus` lists it under `recentSessions`, and `KeployCapture.getSessionId()` returns it. Its flows follow within a minute or two. +- **Poll `uiCaptureStatus` at most once a minute.** Flows are extracted every minute. The state is `FLOWS_READY` as soon as one flow exists, while later visits may still be processing: `sessionsByStatus` says when every one is `PROCESSED`. No state lasts forever: a visit that cannot be processed is retried for about two hours and then counted as failed. +- **The journey plan:** a read of it computes a new one once enough sessions have been processed since the last (`freshness.recomputesAfterSessions`), and once the last is a day old (`freshness.recomputesAt`). `uiJourneys` with `refresh: true` computes a new one from `freshness.refreshableAt`, a minute after the last; before then it answers the stored plan. `refresh` needs `testgen:run`. +- **A UI test run that is `PENDING`** has not been claimed by a runner, and none may claim it: its `next` says not to poll for it. +- **Rate limits.** Each API key may make 100 requests a minute. Past that, the answer is `429` `RATE_LIMITED`. Over REST, its `Retry-After` header says how many seconds to wait. An MCP tool's answer carries no headers: a refused `uiJourneys` says in its message to retry in 10 seconds, and the rate limit's message says only to retry later, so wait a minute. + +## Errors an agent should handle {#errors-an-agent-should-handle} + +The REST API answers an error as `{"error": {"code", "message", "details"}, "meta": {...}}`. An MCP tool returns the same body with `isError` set. Branch on `code`, never on the message. + +| HTTP | `code` | When | What to do | +| ----- | ------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | ------------------------------------------------------------------------------------------------------------------------ | +| `400` | `VALIDATION_ERROR` | A request the API's schema or the route refuses: a key name over 100 bytes or the reserved `Web UI Session`, `ttlDays` outside 1 to 365, a `listUIFlows` `limit` outside 1 to 200, more than 100 `journeyIds`, a malformed ID. | Fix the input: the message says what is wrong. | +| `400` | `INVALID_PARAMETER` | A `uiJourneyImpact` journey that is not one of the plan's `journeys`. | Read the plan again and pass its `journeys`. | +| `401` | `AUTHENTICATION_REQUIRED` | No key, or a key the server does not accept. | Ask the person for a new key or sign-in. | +| `403` | `INSUFFICIENT_SCOPE` | The key lacks the operation's scope. An ingest key gets this on every REST route. | Use a key with the scope; the [REST API](./rest-api.md) lists each one's. | +| `403` | `INSUFFICIENT_PERMISSION` | The key's owner lacks the permission on the app. The message names it: `testgen:run` or `app:view`. | Ask an admin for that permission on the app. | +| `403` | `FORBIDDEN` | A service-account key asked for what no service account may do, such as `listUIIngestKeys`. For `createApp` and `deleteApp`, a `UI_READER` key gets `INSUFFICIENT_SCOPE` first. | Do not retry: a person's key does these. | +| `404` | `RESOURCE_NOT_FOUND` | The app is not the caller's, or the flow, key or run is not this app's. `uiJourneyImpact` with a plan older than 14 days. | Check the IDs; read the plan again for a current `planId`. | +| `409` | `UI_INGEST_KEY_ACCOUNT_CLAIMED` | Minting a key lost a race with the app's service account being torn down, and the key was taken back. | Call `uiCaptureSetup` again. | +| `429` | `RATE_LIMITED` | The key's rate limit, or `uiJourneys` while the server's queue of plan computations is full or the request ended before its plan was ready. | Wait, then retry: `Retry-After` seconds over REST; over MCP, the 10 seconds a refused `uiJourneys` names, else a minute. | +| `501` | `FEATURE_NOT_ENABLED` | The server runs without the feature. | Do not retry. | + +The capture SDK reports its own errors in the browser, each with a `KUI_E_` code: see [UI Capture errors and warnings](./errors.md). + +## What an agent sees {#what-an-agent-sees} + +Flows returned by the REST API and MCP are meant for an agent's context, so they return the shape of the journey and leave out what is most likely to carry your users' data: + +- A step says whether the user entered a value (`hasValue`), never the value itself. +- Request and response headers and bodies are not returned. +- URL query strings, fragments and credentials (`user:password@`) are removed. +- A path segment that looks like a secret or a personal number is replaced with `REDACTED`: an email address, six or more digits (a one-time code, an account or card number, and also a date written as digits, such as `20240115`), twelve or more letters and digits in one run (an invite or reset token), or a long token that is not a UUID. Long IDs that are not UUIDs, such as 24-character MongoDB ObjectIds, are replaced too. +- A flow's name and description are scrubbed the same way, and an `aria-label` value in a selector is replaced with `REDACTED`, because it is page text. + +What is kept as recorded: the page's paths otherwise, and the `data-testid` values, IDs and class names your developers wrote. These checks go by shape, so they can't recognise every piece of personal data: if your app puts user data into paths or into those attributes, it reaches the agent. + +The dashboard shows more to people on your team who can view the app: a flow's recorded URLs with their queries, for example. + +## What agents cannot do yet {#what-agents-cannot-do-yet} + +- **Generate or run tests.** No tool or REST route creates, claims or completes a UI test run. A run today comes only from a flow page's **Generate Test** in the dashboard, and it stays pending: the runner that replays journeys as Playwright tests, `keploy ui run`, is not in a release build yet. +- **Pin a journey** into the plan, or accept a change. +- **Run tests in CI.** No service-account role runs UI tests. A CI job can check capture status and read the plan with a `UI_READER` key, on a paid plan. + +## Related + +- [AI agent quickstart](./quickstart-agent.md) +- [An agent onboards a web app](./walkthrough-agent.md) +- [MCP tools](./mcp-tools.md) +- [REST API](./rest-api.md) +- [CLI](./cli.md) diff --git a/versioned_docs/version-4.0.0/ui-capture/cli.md b/versioned_docs/version-4.0.0/ui-capture/cli.md new file mode 100644 index 000000000..fbae3d53e --- /dev/null +++ b/versioned_docs/version-4.0.0/ui-capture/cli.md @@ -0,0 +1,97 @@ +--- +id: cli +title: "UI Testing CLI reference: keploy login and keploy mcp-install" +sidebar_label: CLI +description: What the keploy binary does for UI Testing today, signing in and connecting an AI agent to the Keploy MCP server, and what it does not do yet. +tags: + - UI testing + - UI capture + - CLI + - reference +keywords: + - keploy login + - keploy mcp-install + - Keploy CLI + - AI agent + - MCP + - UI capture +--- + +import ProductTier from '@site/src/components/ProductTier'; + + + +The `keploy` binary has no UI Testing commands of its own. For UI Testing it does two things: it signs you in to Keploy, from a terminal or from an AI agent's shell, and it connects your AI agent to the Keploy MCP server, whose tools then set up capture and read flows and journeys. This page covers the released `keploy` binary. + +To install it, see [Install Keploy](../server/installation.md). + +## Sign in from a terminal or an agent {#keploy-login} + +```bash +keploy login +``` + +`keploy login` signs you in through your browser and waits until you approve the sign-in. With an API key in `KEPLOY_API_KEY`, or a session it saved before, it uses that instead. + +An AI agent's shell shows a command's output only when the command exits, so an agent signs in in two steps: + +```bash +keploy login --no-wait --json +# a person opens the link and approves the code +keploy login --finish --json +``` + +1. `keploy login --no-wait` prints the sign-in link and code and exits without waiting. The agent shows them to you. +2. You open the link and approve the code. +3. `keploy login --finish` collects the session. It waits up to 70 seconds for the approval, and can be run again until the code expires. + +| Flag | What it does | +| ---------------- | ------------------------------------------------------------------------------------------------------------------------------------- | +| `--no-wait` | Print the sign-in link and code and exit instead of waiting. | +| `--finish` | Collect the session for the sign-in started by `--no-wait`. | +| `--new` | With `--no-wait`: discard the sign-in waiting here and start a new one. Without it, `--no-wait` shows the waiting code or link again. | +| `--json` | With `--no-wait` or `--finish`: print the answer as one JSON document on stdout, for AI agents and tools. Logs go to stderr. | +| `--api-key` | A Keploy API key for this invocation, in place of the `KEPLOY_API_KEY` environment variable. | +| `--show` | Print the API key a login uses here (its last 4 characters only) and where it comes from, then exit without signing in. | +| `--manual-login` | Ask for an API key to paste instead of opening the browser. | + +When an API key is set, in `KEPLOY_API_KEY` or with `--api-key`, and the server accepts it, `keploy login --no-wait` answers that you are signed in and starts no sign-in. + +## Connect your agent {#keploy-mcp-install} + +```bash +keploy mcp-install --editor claude-code +``` + +`keploy mcp-install` uses the session `keploy login` saved: sign in first. It mints a personal access token for the Keploy MCP server and writes the server, `https://api.keploy.io/client/v1/mcp` with that token, into your agent's MCP configuration as `keploy`. The UI capture tools are then available to the agent: see [MCP tools](./mcp-tools.md). + +| Flag | What it does | +| ------------------ | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | +| `--editor ` | The agent to configure: `claude-code`, `cursor`, `windsurf`, `claude-desktop`, `codex`, `antigravity` or `github-copilot`. For `github-copilot`, which has no configuration file on disk, it prints a block to paste. Pass this or `--all`. | +| `--all` | Configure every MCP-aware agent found on this machine. GitHub Copilot is not detected: configure it with `--editor github-copilot`. | +| `--scope ` | `global` (the default) configures the agent for every project on this machine. `project` configures it for the current directory, for Claude Code, Cursor and Codex; for Cursor and Codex that writes the token into a file in your repo. | +| `--api-server-url` | The Keploy API server to connect to. The default is `https://api.keploy.io`. Pass your own server's URL when you self-host. | +| `--rotate` | Mint a new token and revoke the current one. Every agent configured with the old token must then be configured again. | +| `--show-token` | Print the real token inside the copy-paste blocks. Off by default, because those blocks end up in terminal scrollback and CI logs. | + +The token is minted once and reused: `mcp-install` keeps it in `~/.keploy/mcp.yaml`, so configuring one more agent leaves the others working. It expires after 90 days. When your agents start getting `401`, run `keploy mcp-install --rotate`: without `--rotate`, `mcp-install` hands out the cached token again. Minting a new one, with `--rotate`, on another machine or by signing in to the Keploy VS Code extension, which mints its token the same way, revokes the ones minted before, and agents still holding them lose access. + +The token carries the scopes your role in the company allows. Effective access is still what your role holds on each app: setting up capture needs `testgen:run` on the app, and reading flows needs `app:view`. + +## In CI {#in-ci} + +No person can approve a sign-in code in CI. To sign the `keploy` binary in there, set `KEPLOY_API_KEY` to a Keploy API key. + +The UI capture steps a CI job takes, such as checking capture status or reading the journey plan, are calls to the [REST API](./rest-api.md) with an API key, not `keploy` commands. + +## keploy ui {#keploy-ui} + +The `keploy ui` commands, the runner `keploy ui run` that replays journeys as Playwright tests among them, are not in a release build yet. In the released binary, `keploy ui --help` answers `Error: unknown command "ui" for "keploy"`. Replaying journeys as Playwright tests is in beta: see [what works today](../running-keploy/ui-capture.md#what-works-today). + +## Related + +- [AI agent quickstart](./quickstart-agent.md) +- [Operate UI capture from an AI agent](./agents.md) +- [MCP tools](./mcp-tools.md) +- [REST API](./rest-api.md) +- [Install Keploy](../server/installation.md) diff --git a/versioned_docs/version-4.0.0/ui-capture/errors.md b/versioned_docs/version-4.0.0/ui-capture/errors.md index 7deeea22b..22949f889 100644 --- a/versioned_docs/version-4.0.0/ui-capture/errors.md +++ b/versioned_docs/version-4.0.0/ui-capture/errors.md @@ -124,7 +124,7 @@ A program that reads the console can match the code and parse the JSON after the **Fix:** Mint a UI ingest key for this app and pass it as `apiKey` to `KeployCapture.init()`. You can mint one: - In the Keploy dashboard, as described in [Create an ingest key](./quickstart-dashboard.md#create-an-ingest-key) in the dashboard quickstart. -- With `POST /client/v1/apps//ui/ingest-keys` and a scoped Public API key with `write` scope in `Authorization: Bearer`, as described in [Create an ingest key](../running-keploy/ui-capture.md#create-an-ingest-key). The Keploy MCP server's `uiCaptureSetup` tool makes the same call. +- With `POST /client/v1/apps//ui/ingest-keys` and a scoped Public API key with `write` scope in `Authorization: Bearer`, as described in [Create an ingest key](./rest-api.md#create-an-ingest-key). The Keploy MCP server's `uiCaptureSetup` tool makes the same call. Both need `testgen:run` on the app. A new key takes effect on the next page load. @@ -525,7 +525,7 @@ How to act on one, as a person or as an AI agent: 3. **Read the exit code before you read a run as a test result.** Only `0` and `1` are results of a run. Every other code means nothing, or nothing trustworthy, was measured. 4. **A code marked reserved is not raised by any current build.** It documents a check that is designed but not wired yet. Do not wait for it or branch on it. -A remedy names only these commands of the keploy binary: `keploy login` with `--no-wait`, `--json` and `--finish`, `keploy ui --help`, and `keploy ui run`, which needs `--app` and `--base-url` and also takes `--suite`, `--plan`, `--repo` and `--out`. Any other command a remedy names, such as `npx playwright test`, is not the binary's. +A remedy names only these commands of the keploy binary: `keploy login` with `--no-wait`, `--json` and `--finish`, which the released binary has, and `keploy ui --help` and `keploy ui run`, which are not in a release build yet. Any other command a remedy names, such as `npx playwright test`, is not the binary's. ### Exit codes diff --git a/versioned_docs/version-4.0.0/ui-capture/mcp-tools.md b/versioned_docs/version-4.0.0/ui-capture/mcp-tools.md new file mode 100644 index 000000000..3116ff863 --- /dev/null +++ b/versioned_docs/version-4.0.0/ui-capture/mcp-tools.md @@ -0,0 +1,124 @@ +--- +id: mcp-tools +title: "UI Testing MCP tools reference" +sidebar_label: MCP tools +description: The Keploy MCP server's UI capture tools, their inputs, the scope and app permission each needs, and how an agent finds and calls them. +tags: + - UI testing + - UI capture + - MCP + - reference +keywords: + - Keploy MCP server + - MCP tools + - uiCaptureSetup + - uiCaptureStatus + - AI agent + - invoke_tool +--- + +import ProductTier from '@site/src/components/ProductTier'; + + + +The Keploy MCP server gives an AI agent every step of UI capture as a tool: create an app, mint an ingest key, check that recordings arrive, and read flows, journeys and test runs. Each UI capture tool is generated from the [REST API](./rest-api.md) operation of the same name, so it takes the same inputs and returns the same answer. + +## The server {#the-server} + +| What | Value | +| -------------- | -------------------------------------------------------------------------------------------------------------------------------------------- | +| URL | `https://api.keploy.io/client/v1/mcp`. When you self-host, your API server's URL followed by `/client/v1/mcp`. | +| Transport | Streamable HTTP, stateless. | +| Server name | `keploy-atg` | +| Authentication | An OAuth sign-in, or a Keploy API key in `Authorization: Bearer kep_...` or `X-API-Key: kep_...`. See [OAuth or API key](#oauth-or-api-key). | +| Rate limit | Each tool call that reaches the API counts against the key's 100 requests a minute, as a REST call does. | + +**Tool-search mode.** The server's `tools/list` shows seven tools, so that the full catalogue does not fill the agent's context: the meta-tools `search_tools`, `get_tool_schema` and `invoke_tool`, and the sign-in and setup tools `get_auth_status`, `get_setup_instructions`, `devloop_setup_instructions` and `devloop_begin_oauth_install`. Every other tool is still there: + +- `get_tool_schema({names: [...]})` returns the description and input schema of each tool you name. Ask for all you need in one call: the names are in the [table below](#tools) and in the server's instructions. +- `search_tools({query})` finds tools by keyword and returns each match's name and a one-line summary, not its schema. It ranks the whole catalogue by keyword and returns 10 matches unless you pass `limit`, so a query such as `"ui capture"` can return API testing tools too and leave some UI capture tools out. Fetch tools whose names you know with `get_tool_schema`. +- `invoke_tool({name, arguments})` runs a tool by name. A tool missing from `tools/list` can also be called directly with `tools/call` and its exact name. + +**Instructions.** The server's `initialize` answer carries instructions with a UI CAPTURE section: the order of the tools below, the wait before a visit becomes flows, and that the replay runner is not in a release build yet, so an agent should not tell you to run it. + +## Tools {#tools} + +Every UI capture tool takes `appId`, the app's ID. The scope is the API key's; the permission is what the key's owner must hold on the app. + +| Tool | What it does | Inputs | Scope and permission | +| ------------------- | ---------------------------------------------------------------------------------------------------------- | ----------------------------------------------------- | ------------------------------------------------------------------------ | +| `createApp` | Creates an app. An app for UI capture needs only a `name`, which cannot be changed later. | `name` (required), and other optional app settings | `write` | +| `listApps` | Lists the company's apps; `q` filters by name. | `q`, `offset`, `limit` | `read` | +| `getApp` | Returns one app. | `appId`, `fields` | `read` | +| `deleteApp` | Deletes an app. | `appId` | `admin` | +| `uiCaptureSetup` | Mints an ingest key and returns it once, with the endpoint, the SDK install command and an `init` snippet. | `appId`, `name`, `ttlDays` | `write`, `testgen:run` | +| `listUIIngestKeys` | Lists the app's ingest keys, revoked ones included, never their secrets. | `appId` | `read`, `app:view` | +| `revokeUIIngestKey` | Revokes one ingest key. | `appId`, `keyId` | `write`, `testgen:run` | +| `uiCaptureStatus` | Says whether recordings arrive and have become flows, as a state with what to do next. | `appId` | `read`, `app:view` | +| `listUIFlows` | Lists the app's flows. | `appId`, `limit`, `offset`, `status` | `read`, `app:view` | +| `getUIFlow` | Returns one flow's steps, network calls and end state. | `appId`, `flowId` | `read`, `app:view` | +| `uiFlowGraph` | How the app's visits move between pages. | `appId` | `read`, `app:view` | +| `uiJourneys` | The journey plan: the journeys sessions take, each with its share of sessions. | `appId`, `refresh` | `read`, `app:view`; `refresh: true` also needs `write` and `testgen:run` | +| `uiJourneyImpact` | The share of sessions that took at least one of a set of the plan's journeys. | `appId`, `planId` (required), `journeyIds` (required) | `read`, `app:view` | +| `listUITestRuns` | Lists the app's UI test runs, newest first, one summary line each. | `appId`, `offset`, `limit` | `read`, `app:view` | +| `getUITestRun` | Returns one UI test run with each journey's verdict. | `appId`, `runId` | `read`, `app:view` | + +The [REST API reference](./rest-api.md) describes each tool's answer in full. A UI test run today comes only from a flow page's **Generate Test** in the dashboard, and it stays pending: no tool creates or runs one, and the runner that replays journeys is not in a release build yet. + +## Call a tool {#call-a-tool} + +An MCP client makes these calls for your agent. To see one by hand, send JSON-RPC to the server: + +```bash +curl https://api.keploy.io/client/v1/mcp \ + -H "Authorization: Bearer kep_YOUR_API_KEY" \ + -H "Content-Type: application/json" \ + -H "Accept: application/json, text/event-stream" \ + -d '{ + "jsonrpc": "2.0", + "id": 1, + "method": "tools/call", + "params": { + "name": "invoke_tool", + "arguments": {"name": "uiCaptureStatus", "arguments": {"appId": ""}} + } + }' +``` + +The tool's answer is one text item: the REST API's JSON answer, with the HTTP status added as `status_code`. A refusal sets the result's `isError`, and its text carries the API's `error`, as here for `getUITestRun` with a run that does not exist: + +```json +{ + "error": { + "code": "RESOURCE_NOT_FOUND", + "message": "UI test run not found" + }, + "meta": { + "timestamp": "2026-11-20T07:28:18Z" + }, + "status_code": 404 +} +``` + +## Input types {#input-types} + +A REST operation's path and query parameters become `string`, `number` or `boolean` inputs, and its request body's fields keep their types. Two cases to know: + +- **`journeyIds` is one string**, the journey IDs joined with commas: `"journeyIds": "a1b2…,c3d4…"`. It takes 1 to 100 IDs, each one of the plan's `journeys`. +- **`refresh` is a boolean**: `"refresh": true`. + +## OAuth or API key {#oauth-or-api-key} + +- **OAuth.** A request with no credential is answered `401` with a `WWW-Authenticate` challenge that names the server's OAuth metadata, and an MCP client that supports OAuth signs in from it: you approve the sign-in in your browser. The token it gets has the `read` and `write` scopes, never `admin`, so it cannot call `deleteApp`. It works only through the MCP server, not on the REST API. +- **An API key.** A personal access token from **Settings → API Keys** in the dashboard, or the one [`keploy mcp-install`](./cli.md#keploy-mcp-install) mints, in `Authorization: Bearer kep_...`. Give it the `read` and `write` scopes to set up capture; deleting an app needs `admin`, which only organization admins can give a token. +- **A service-account key** bound to the `UI_READER` role on an app reads that app's capture status, flows, flow graph, journeys, impact and test runs. It cannot create or delete apps, list ingest keys, or mint or revoke them. On the Free plan a service account can hold only `MOCK_READER`, so this needs a paid plan. + +An ingest key is not accepted: it can only post recordings. + +## Related + +- [AI agent quickstart](./quickstart-agent.md) +- [An agent onboards a web app](./walkthrough-agent.md) +- [Operate UI capture from an AI agent](./agents.md) +- [REST API](./rest-api.md) +- [CLI](./cli.md) diff --git a/versioned_docs/version-4.0.0/ui-capture/privacy.md b/versioned_docs/version-4.0.0/ui-capture/privacy.md index 8de07a04f..a734a5fa4 100644 --- a/versioned_docs/version-4.0.0/ui-capture/privacy.md +++ b/versioned_docs/version-4.0.0/ui-capture/privacy.md @@ -212,7 +212,7 @@ Each request also carries the ingest key in its `X-API-Key` header and a random ## What the dashboard, REST and MCP show {#what-the-dashboard-rest-and-mcp-show} - **The dashboard** shows a flow to everyone who can view its app. A flow's page lists its steps, with a typed value as it was recorded (a masked one as asterisks), and its network calls with their method, recorded URL, status and time. The page shows no headers or bodies, but Keploy stores each call's headers and bodies, as the SDK recorded them, with its flow, because a replay serves them, and anyone who can view the app can read them through the dashboard's own API. -- **The REST API and MCP** return less, because their answers go into an AI agent's context: no typed values, no headers or bodies, no query strings, fragments or credentials in URLs, and a path segment that reads as a secret or a person's data replaced with `REDACTED`. [What an agent sees](../running-keploy/ui-capture.md#what-an-agent-sees) lists the rules. +- **The REST API and MCP** return less, because their answers go into an AI agent's context: no typed values, no headers or bodies, no query strings, fragments or credentials in URLs, and a path segment that reads as a secret or a person's data replaced with `REDACTED`. [What an agent sees](./agents.md#what-an-agent-sees) lists the rules. ## Recognition, not a guarantee {#recognition-not-a-guarantee} diff --git a/versioned_docs/version-4.0.0/ui-capture/quickstart-agent.md b/versioned_docs/version-4.0.0/ui-capture/quickstart-agent.md new file mode 100644 index 000000000..5d0da9ae0 --- /dev/null +++ b/versioned_docs/version-4.0.0/ui-capture/quickstart-agent.md @@ -0,0 +1,176 @@ +--- +id: quickstart-agent +title: "UI Testing quickstart for AI agents: set up capture from Claude Code, Cursor or Codex" +sidebar_label: AI agent quickstart +description: Connect your AI coding agent to the Keploy MCP server and have it set up UI capture in your web app and read the first flows for you. +tags: + - UI testing + - UI capture + - quickstart + - AI agent + - MCP +keywords: + - UI capture quickstart + - AI agent + - Keploy MCP server + - Claude Code + - Cursor + - Codex +--- + +import HowTo from '@site/src/components/HowTo'; +import ProductTier from '@site/src/components/ProductTier'; + +/complete, or waits 30 minutes.", +url: "#use-your-app", +}, +{ +name: "Read the flows", +text: "When uiCaptureStatus says FLOWS_READY, the agent reads the flows with listUIFlows and getUIFlow, and the journeys with uiJourneys.", +url: "#read-the-flows", +}, +]} +/> + + + +In this quickstart your AI coding agent sets up Keploy UI capture for you. You connect it to the Keploy MCP server and ask it to set up capture; the agent mints an ingest key, adds the capture SDK to your frontend, and reads the **UI flows** your first visits make. Apart from creating a token, if you connect with one, you never open the dashboard. It takes about 10 minutes. + +:::note Beta +UI Testing is in beta. Recording visits and reading them as flows and journeys works today. Replaying journeys as Playwright tests is in beta too, and its runner is not in a release build yet, so this quickstart stops at the flows. See [what works today](../running-keploy/ui-capture.md#what-works-today). +::: + +## Before you begin + +- A Keploy account. Sign in at [app.keploy.io](https://app.keploy.io). +- An AI coding agent that can use MCP servers over HTTP, such as Claude Code, Cursor, Codex, Windsurf or Antigravity. +- Your web app's repository, open in the agent. To try it on a sample first, use the [React sample shop](https://github.com/keploy/samples-typescript/tree/main/ui-capture-react-shop): `git clone https://github.com/keploy/samples-typescript`, then open `samples-typescript/ui-capture-react-shop`. +- Permission to run test generation on the app (`testgen:run`), and a credential with the `write` scope: minting an ingest key needs both, and creating an app needs `write`. + +## 1. Connect your agent to Keploy {#connect-your-agent} + +The Keploy MCP server is at `https://api.keploy.io/client/v1/mcp`. Connect your agent to it one of three ways. + +**Sign in with OAuth.** Add the server with no key. For Claude Code: + +```bash +claude mcp add --transport http keploy https://api.keploy.io/client/v1/mcp +``` + +The server answers a request with no key with `401` and a pointer to its OAuth metadata, and an agent that supports MCP OAuth starts the sign-in from that by itself: you approve it in your browser. The sign-in can read and write, but not delete apps. It is listed under **Settings → Authorized apps** in the dashboard, where you can revoke it. + +**With a personal access token.** In the dashboard, open **Settings → API Keys → Generate new token**, and give the token the **Read** and **Write** scopes. Put it in `KEPLOY_PAT`, then add the server with it. For Claude Code: + +```bash +claude mcp add --transport http keploy https://api.keploy.io/client/v1/mcp \ + --header "Authorization: Bearer $KEPLOY_PAT" +``` + +This is the command the setup card under **UI Testing → UI Flows** shows, for the server your dashboard uses: + +The setup card's section Setting this up with an AI agent or a script?, with the claude mcp add command for the Keploy MCP server and a curl command that mints an ingest key + +**With the keploy CLI.** If you have the `keploy` binary, sign in and let it configure your agent: + +```bash +keploy login +keploy mcp-install --editor claude-code +``` + +`--editor` also takes `cursor`, `windsurf`, `claude-desktop`, `codex`, `antigravity` and `github-copilot`. [CLI](./cli.md) has every flag, and how an agent runs `keploy login` without waiting for you. + +Other MCP clients take the same URL and header in their configuration file: see [MCP client configuration](../running-keploy/agent-test-generation.md#mcp-client-configuration). + +## 2. Ask your agent {#ask-your-agent} + +In your app's repository, ask the agent, for example: + +> "Use the Keploy MCP tools to set up UI capture for this frontend. Create a Keploy app named `storefront-web` if I don't have one, mint an ingest key, add the capture SDK so it starts before the app's first request, and tell me when the first flows are ready." + +## 3. What the agent does {#what-the-agent-does} + +The server runs in tool-search mode, so its UI capture tools are not in the agent's initial tool list. The server's own instructions name them, so the agent fetches their schemas in one `get_tool_schema` call and runs them with `invoke_tool`. Then: + +1. **`listApps`** finds your app, or **`createApp`** creates one. An app for UI capture needs only a name. +2. **`uiCaptureSetup`** mints an ingest key for the app. The answer holds the key, which it shows only once, the ingest endpoint, the install command `npm install @keploy/capture-sdk`, and an `init` snippet with the endpoint and key in it. +3. **The agent edits your frontend.** It installs the SDK and calls `KeployCapture.init` once, in a module your app imports before anything else. The ingest key can only post recordings into this one app, so it is safe in your page's JavaScript. In the sample shop, it puts the key and endpoint in `.env`, which `src/keploy.ts` reads. The snippet leaves `privacyMode` at its default, `"balanced"`, which masks everything visitors type; for a public site, ask the agent to set `"strict"`, which masks all visible text too. +4. **`uiCaptureStatus`** says `NO_SESSIONS` until someone uses the app. + +## 4. Use your app {#use-your-app} + +Start your app and use it in a browser for a minute: open a few pages, search, click. `uiCaptureStatus` then says `RECORDING`. + +A visit becomes flows 30 minutes after its last activity. To see flows sooner, close the app's tab and tell the agent: it ends the visit at once with `POST /ui/v1/sessions//complete` and the ingest key, and the flows follow within a minute or two. The session id is the visit's `sdkSessionId`, which `uiCaptureStatus` lists under `recentSessions`; it is also what `KeployCapture.getSessionId()` returns, and the sample shop prints it in the browser console. The agent checks `uiCaptureStatus` no more than once a minute until it says `FLOWS_READY`. + +## 5. Read the flows {#read-the-flows} + +The agent reads what your visit recorded: + +- **`listUIFlows`** lists the flows, usually one for each page a visit opened, named after what happened and where, such as `Fill on /products`. +- **`getUIFlow`** returns one flow's steps and the network calls they made. +- **`uiJourneys`** returns the journey plan: the paths your sessions took through your pages, each with its share of sessions. + +[An agent onboards a web app](./walkthrough-agent.md) shows each of these calls with the answer it returns. + +## Without MCP: the same steps with curl {#without-mcp} + +An agent or a script without MCP makes the same calls on the [REST API](./rest-api.md), with a personal access token: + +```bash +API=https://api.keploy.io/client/v1 +AUTH="Authorization: Bearer $KEPLOY_PAT" + +# Create the app (skip this for an existing one) +curl -X POST "$API/apps" -H "$AUTH" -H "Content-Type: application/json" -d '{"name": "storefront-web"}' + +# Mint an ingest key: the answer has the key, the endpoint and the init snippet +curl -X POST "$API/apps//ui/ingest-keys" -H "$AUTH" -H "Content-Type: application/json" -d '{"name": "storefront web"}' + +# Check capture, read the flows and the journey plan +curl "$API/apps//ui/capture-status" -H "$AUTH" +curl "$API/apps//ui/flows" -H "$AUTH" +curl "$API/apps//ui/journeys" -H "$AUTH" +``` + +## What the agent sees {#what-the-agent-sees} + +Flows returned to an agent carry the shape of each visit, not your users' data: a step says that a value was typed, never what; headers, bodies and URL queries are left out; and a path segment that looks like an email address, a long number or a token comes back as `REDACTED`. [What an agent sees](./agents.md#what-an-agent-sees) lists the rules. + +## Clean up {#clean-up} + +To stop recording, ask the agent to revoke the ingest key: it lists the app's keys with `listUIIngestKeys` and revokes one with `revokeUIIngestKey`. Pages that still use the key stop recording within about 30 seconds. + +Deleting an app needs a personal access token with the **Admin** scope, which only organization admins can give. An OAuth sign-in cannot delete apps. To disconnect an agent that signed in with OAuth, revoke it under **Settings → Authorized apps**. + +## Related + +- [An agent onboards a web app](./walkthrough-agent.md) +- [Operate UI capture from an AI agent](./agents.md) +- [MCP tools](./mcp-tools.md) +- [CLI](./cli.md) +- [Dashboard quickstart](./quickstart-dashboard.md) diff --git a/versioned_docs/version-4.0.0/ui-capture/rest-api.md b/versioned_docs/version-4.0.0/ui-capture/rest-api.md new file mode 100644 index 000000000..5b7cf8e09 --- /dev/null +++ b/versioned_docs/version-4.0.0/ui-capture/rest-api.md @@ -0,0 +1,737 @@ +--- +id: rest-api +title: "UI Testing REST API reference" +sidebar_label: REST API +description: Every UI capture operation of the Keploy Public API, from minting ingest keys to capture status, flows, journeys and test runs, with real answers. +tags: + - UI testing + - UI capture + - REST API + - reference +keywords: + - Keploy Public API + - UI capture API + - ingest key + - capture status + - UI flows + - journeys +--- + +import ProductTier from '@site/src/components/ProductTier'; + + + +The UI capture operations of the [Keploy Public API](../running-keploy/public-api.md) let a script or an AI agent do everything the dashboard's UI Testing pages do: mint the ingest key the capture SDK sends with, check that recordings arrive, and read flows, journeys and test runs. Each operation is also a tool of the same name on the Keploy MCP server: see [MCP tools](./mcp-tools.md). + +The answers on this page are real: they were captured from an app with six recorded visits to the [React sample shop](https://github.com/keploy/samples-typescript/tree/main/ui-capture-react-shop), some through the REST API and some through the MCP server, which returns the same answer. IDs are replaced, `meta.request_id` and `meta.trace_id` are left out, and long lists are shortened. + +## Base URL and authentication {#base-url-and-authentication} + +- **Base URL:** `https://api.keploy.io/client/v1`. When you self-host, your API server's URL followed by `/client/v1`. +- **API key:** a personal access token from **Settings → API Keys** in the dashboard, in `Authorization: Bearer kep_...` or `X-API-Key: kep_...`. An ingest key is refused with `403`: it can only post recordings. +- **Scope and permission.** Each operation needs a scope on the key (`read`, `write` or `admin`) and a permission its owner holds on the app: `app:view` to read, `testgen:run` to mint or revoke an ingest key. A missing scope is `403` `INSUFFICIENT_SCOPE`; a missing permission is `403` `INSUFFICIENT_PERMISSION`, with the permission named. +- **Service accounts.** A service-account key bound to the `UI_READER` role on an app may call that app's read operations, except listing its ingest keys. It may not create or delete apps, or mint or revoke ingest keys. On the Free plan a service account can hold only `MOCK_READER`, so this needs a paid plan. +- **Rate limit:** 100 requests a minute per key, with `X-RateLimit-Limit`, `X-RateLimit-Remaining` and `X-RateLimit-Reset` on each answer, and `429` `RATE_LIMITED` with `Retry-After` past it. +- **Answers** are `{"data": ..., "meta": {"request_id", "timestamp"}}`. Errors are `{"error": {"code", "message", "details"}, "meta": ...}`. Branch on `code`: [Error codes](../running-keploy/public-api.md#error-codes) lists the Public API's, and each operation below names the ones it adds. +- **Times** in `data` are Unix seconds, except the `timestamp` of a flow's steps and calls, which is in milliseconds. `meta.timestamp` is an RFC 3339 string. +- **The spec** is served at `GET /client/v1/openapi.yaml`, and as a page at `/client/v1/docs`, without a key. The UI capture operations carry the tag `UI Capture`. + +## Create an app {#create-an-app} + +`POST /client/v1/apps` · MCP `createApp` · scope `write` + +An app for UI capture needs only a `name`. The name cannot be changed later, and two apps cannot share one. + +```bash +curl -X POST https://api.keploy.io/client/v1/apps \ + -H "Authorization: Bearer kep_YOUR_API_KEY" \ + -H "Content-Type: application/json" \ + -d '{"name": "storefront-web"}' +``` + +The answer, `201`, is the app; its `id` is the `appId` every operation below takes. + +## Create an ingest key {#create-an-ingest-key} + +`POST /client/v1/apps/{appId}/ui/ingest-keys` · MCP `uiCaptureSetup` · scope `write`, permission `testgen:run` + +Mints an ingest key for the app and returns everything needed to install the capture SDK. + +| Body field | Type | What it is | +| ---------- | ------- | --------------------------------------------------------------------------------------------------------------------------------------------------------- | +| `name` | string | Optional. A label for the key, at most 100 bytes, spaces at either end trimmed. Without one, `ui-capture` and today's date. `Web UI Session` is reserved. | +| `ttlDays` | integer | Optional. The key's lifetime in days, 1 to 365. Without it the key does not expire. | + +```bash +curl -X POST "https://api.keploy.io/client/v1/apps//ui/ingest-keys" \ + -H "Authorization: Bearer kep_YOUR_API_KEY" \ + -H "Content-Type: application/json" \ + -d '{"name": "ci smoke"}' +``` + +```json title="201 Created" +{ + "data": { + "key": { + "id": "0a09758e-d08e-4354-913f-dfb79fbb52bc", + "name": "ci smoke", + "keyPrefix": "kep_YOUR_ING", + "appId": "af1ec8ee-c43a-4c9c-826b-4223f506fb19", + "status": "active", + "createdAt": 1795159504 + }, + "apiKey": "kep_YOUR_INGEST_KEY", + "endpoint": "https://api.keploy.io/ui/v1/ingest", + "sdk": { + "package": "@keploy/capture-sdk", + "install": "npm install @keploy/capture-sdk", + "init": "import { KeployCapture } from \"@keploy/capture-sdk\";\n\nawait KeployCapture.init({\n endpoint: \"https://api.keploy.io/ui/v1/ingest\",\n apiKey: \"kep_YOUR_INGEST_KEY\",\n});\n" + }, + "next": [ + "apiKey is shown only in this response: store it now (it is safe to ship in the page; it can only post recordings into this app)", + "install the SDK and call KeployCapture.init once at app start-up, in the browser bundle", + "browse the app, then GET /client/v1/apps/af1ec8ee-c43a-4c9c-826b-4223f506fb19/ui/capture-status (MCP: uiCaptureStatus) to see sessions arrive", + "a session becomes flows 30 minutes after its last activity (to end one at once, as a test or CI recording does, POST https://api.keploy.io/ui/v1/sessions/{sdkSessionId}/complete with this key in X-API-Key); read them with GET /client/v1/apps/af1ec8ee-c43a-4c9c-826b-4223f506fb19/ui/flows (MCP: listUIFlows)", + "once sessions are processed, GET /client/v1/apps/af1ec8ee-c43a-4c9c-826b-4223f506fb19/ui/journeys (MCP: uiJourneys) is the journey plan a UI test run replays; GET /client/v1/apps/af1ec8ee-c43a-4c9c-826b-4223f506fb19/ui/test-runs (MCP: listUITestRuns, getUITestRun) reads the runs" + ] + }, + "meta": { + "timestamp": "2026-11-20T07:25:04Z" + } +} +``` + +- `apiKey` is in this answer only, which carries `Cache-Control: no-store`: store it now. It is safe to ship in your page's JavaScript, because it can only post recordings into this one app. +- Every call mints a new key, since a stored key's secret cannot be returned again. List and revoke the ones you no longer use. +- `sdk.init` sets only `endpoint` and `apiKey`. Put it in your app's browser entry point, once, in a module imported before any data fetching: requests made before `init` has loaded rrweb are not recorded. +- `endpoint` is empty when a self-hosted server does not know its public URL (`API_SERVER_URL` is not an absolute URL), and `next` then says to use the base URL you called this API on, followed by `/ui/v1/ingest`. +- A `409` with `UI_INGEST_KEY_ACCOUNT_CLAIMED` means the mint lost a race with a teardown of the app's service account, and the key was taken back. Call it again. + +## List ingest keys {#list-ingest-keys} + +`GET /client/v1/apps/{appId}/ui/ingest-keys` · MCP `listUIIngestKeys` · scope `read`, permission `app:view` + +The app's ingest keys, newest first, revoked ones included, and never their secrets. `keyPrefix` matches the start of the `apiKey` a page ships. + +```bash +curl "https://api.keploy.io/client/v1/apps//ui/ingest-keys" \ + -H "Authorization: Bearer kep_YOUR_API_KEY" +``` + +```json +{ + "data": [ + { + "id": "0a09758e-d08e-4354-913f-dfb79fbb52bc", + "name": "ci smoke", + "keyPrefix": "kep_YOUR_ING", + "appId": "af1ec8ee-c43a-4c9c-826b-4223f506fb19", + "status": "revoked", + "createdAt": 1795159504 + }, + { + "id": "ec6ebf01-0250-49c5-b71c-554cfec6f0af", + "name": "storefront web", + "keyPrefix": "kep_YOUR_ING", + "appId": "af1ec8ee-c43a-4c9c-826b-4223f506fb19", + "status": "active", + "createdAt": 1795159503, + "expiresAt": 1797751503 + } + ], + "meta": { + "timestamp": "2026-11-20T07:25:05Z" + } +} +``` + +## Revoke an ingest key {#revoke-an-ingest-key} + +`DELETE /client/v1/apps/{appId}/ui/ingest-keys/{keyId}` · MCP `revokeUIIngestKey` · scope `write`, permission `testgen:run` + +```bash +curl -X DELETE "https://api.keploy.io/client/v1/apps//ui/ingest-keys/" \ + -H "Authorization: Bearer kep_YOUR_API_KEY" +``` + +```json +{ + "data": { + "id": "0a09758e-d08e-4354-913f-dfb79fbb52bc", + "revoked": true + }, + "meta": { + "timestamp": "2026-11-20T07:25:04Z" + } +} +``` + +Pages that still ship the key are refused within about 30 seconds. A key that is not one of this app's ingest keys is `404`. + +## Capture status {#capture-status} + +`GET /client/v1/apps/{appId}/ui/capture-status` · MCP `uiCaptureStatus` · scope `read`, permission `app:view` + +Whether the capture SDK is delivering sessions and whether they have become flows, as a `state` to branch on, with an `explanation` and what to do `next`. It also returns the ingest `endpoint`, `idleAfterSeconds` (how long after its last activity a visit becomes flows), the number of sessions in all and by status, when the last batch arrived (`lastBatchAt`), the number of flows (`totalFlows`), and the ten most recent sessions. + +```bash +curl "https://api.keploy.io/client/v1/apps//ui/capture-status" \ + -H "Authorization: Bearer kep_YOUR_API_KEY" +``` + +```json title="While a visit is recording" +{ + "data": { + "state": "RECORDING", + "explanation": "sessions are arriving; a visit becomes flows 30 minutes after its last activity", + "next": [ + "keep using the app, or wait 30 minutes after the last activity, then check again", + "to end a visit at once (a test or CI recording), POST https://api.keploy.io/ui/v1/sessions/{sdkSessionId}/complete with the app's ingest key in X-API-Key (the SDK's own session id for the visit); its flows follow within a minute or two" + ], + "endpoint": "https://api.keploy.io/ui/v1/ingest", + "idleAfterSeconds": 1800, + "totalSessions": 1, + "sessionsByStatus": { + "RECORDING": 1 + }, + "lastBatchAt": 1795159548, + "totalFlows": 0, + "recentSessions": [ + { + "id": "a1e271a4-c6e8-456a-ab38-83f4bf64db4f", + "sdkSessionId": "01056d9a-f9b9-48ef-baf7-15d8804cd7ab", + "status": "RECORDING", + "startedAt": 1795159526, + "lastBatchAt": 1795159548, + "eventCount": 70, + "batches": 5, + "sdkVersion": "3.1.0" + } + ] + }, + "meta": { + "timestamp": "2026-11-20T07:25:49Z" + } +} +``` + +A session's `id` is the `sessionId` its flows carry; `sdkSessionId` is the browser's own id for the visit, the one `POST /ui/v1/sessions//complete` takes. + +| `state` | What it means | What to do | +| ------------- | ---------------------------------------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------- | +| `NO_SESSIONS` | No recording has reached this app. | Check that the page calls `init` with an ingest key for this app and the right endpoint. | +| `RECORDING` | Sessions are arriving. Each becomes flows `idleAfterSeconds` after its last activity. | Wait, keep using the app, or end a visit at once with [`POST /ui/v1/sessions//complete`](#complete-a-session). | +| `PROCESSING` | Finished visits are being turned into flows. | Check again in a minute: extraction runs every minute. | +| `FLOWS_READY` | At least one flow exists. | [List the flows](#list-flows), and read the [journey plan](#journey-plan). | +| `NO_FLOWS` | Visits were processed, and none made a flow. A page opened and left untouched still makes a one-step flow, such as `Navigate on /about`. | Record a fresh visit. | +| `ALL_FAILED` | Every finished visit failed to process. | Record a fresh visit; contact Keploy support if it fails again. | + +A visit that can't be processed is retried with backoff for about two hours and then counted as failed, so no state lasts forever. + +## List flows {#list-flows} + +`GET /client/v1/apps/{appId}/ui/flows` · MCP `listUIFlows` · scope `read`, permission `app:view` + +The app's flows, usually one for each page a visit opened: its name, description, status, the `sessionId` it came from, its number of steps and network calls, and `createdAt`. + +| Query parameter | What it is | +| --------------- | ------------------------------------------------------------------------------------------------------ | +| `limit` | Page size, 1 to 200. The default is 50. | +| `offset` | Page offset. The default is 0. | +| `status` | Only flows in this status: `EXTRACTED`, `CLUSTERED`, `TEST_GENERATED`, `TEST_PASSED` or `TEST_FAILED`. | + +```bash +curl "https://api.keploy.io/client/v1/apps//ui/flows?limit=2" \ + -H "Authorization: Bearer kep_YOUR_API_KEY" +``` + +```json +{ + "data": [ + { + "id": "d07317d4-e2b5-4ba6-b1db-767966e4e474", + "name": "Click on /", + "description": "2 steps, 1 API calls, 6.1s", + "status": "CLUSTERED", + "sessionId": "2a58f279-cc49-48cf-85ab-e7cfd9199a33", + "clusterId": "661be522-1276-4612-9e7e-181dfd1ad3ff", + "steps": 2, + "networkCalls": 1, + "createdAt": 1795159660 + }, + { + "id": "9888b9ee-f61d-4ca3-b13f-f0524c31d4ed", + "name": "Navigate on /about", + "description": "1 steps, 1 API calls, 0.0s", + "status": "CLUSTERED", + "sessionId": "2a58f279-cc49-48cf-85ab-e7cfd9199a33", + "clusterId": "4949ebdb-3873-4f10-9d79-c2027ee534a8", + "steps": 1, + "networkCalls": 1, + "createdAt": 1795159660 + } + ], + "meta": { + "timestamp": "2026-11-20T07:28:15Z" + } +} +``` + +The answer has no `meta.pagination`: ask for the next page with `offset` until a page comes back shorter than `limit`. + +## Get a flow {#get-a-flow} + +`GET /client/v1/apps/{appId}/ui/flows/{flowId}` · MCP `getUIFlow` · scope `read`, permission `app:view` + +One flow in full: what [List flows](#list-flows) says of it, plus `stepList`, `networkCallList` and `outcome`. + +- **`stepList`:** each step's `index`, `action`, `selector`, `hasValue` (a value was entered, never which), `checked` (the state an input step left a checkbox or a radio in), `pageUrl` and `timestamp`. +- **`networkCallList`:** each call's `method`, `url`, `responseStatus` and `timestamp`. +- **`outcome`:** the end state the recording proves: the path the flow's last click led to (`url`), or the elements the app added after its last answered call and still showed, each named by a selector and never by its text; the `viewport`, the window's size when the outcome was read; and `navigation`, how the next flow began. A flow extracted before extraction version 4 has none. + +```bash +curl "https://api.keploy.io/client/v1/apps//ui/flows/" \ + -H "Authorization: Bearer kep_YOUR_API_KEY" +``` + +```json +{ + "data": { + "id": "248a8854-06ac-4189-936b-e7eb2bd55326", + "name": "Fill on /products", + "description": "3 steps, 2 API calls, 6.1s", + "status": "CLUSTERED", + "sessionId": "db452012-c693-4938-a672-c15f44c59b1a", + "clusterId": "16ba6a5d-a195-4ebb-9457-7f697bfc55c7", + "steps": 3, + "networkCalls": 2, + "createdAt": 1795159660, + "stepList": [ + { + "index": 0, + "action": "navigate", + "selector": "", + "hasValue": true, + "pageUrl": "http://localhost:4173/products", + "timestamp": 1795159625548 + }, + { + "index": 1, + "action": "input", + "selector": "[data-testid=\"search-input\"]", + "hasValue": true, + "pageUrl": "http://localhost:4173/products", + "timestamp": 1795159631613 + }, + { + "index": 2, + "action": "click", + "selector": "[data-testid=\"search-button\"]", + "pageUrl": "http://localhost:4173/products", + "timestamp": 1795159631661 + } + ], + "networkCallList": [ + { + "method": "GET", + "url": "http://localhost:4173/api/products", + "responseStatus": 200, + "timestamp": 1795159625571 + }, + { + "method": "GET", + "url": "http://localhost:4173/api/products", + "responseStatus": 200, + "timestamp": 1795159631678 + } + ], + "outcome": { + "elements": [], + "viewport": { + "width": 1280, + "height": 720 + } + } + }, + "meta": { + "timestamp": "2026-11-20T07:28:15Z" + } +} +``` + +Headers, bodies, typed values and URL queries are never returned, and path segments that read as secrets or personal data come back as `REDACTED`: [What an agent sees](./agents.md#what-an-agent-sees) lists the rules. A flow of another app is `404`. + +## Flow graph {#flow-graph} + +`GET /client/v1/apps/{appId}/ui/flow-graph` · MCP `uiFlowGraph` · scope `read`, permission `app:view` + +How the app's recorded visits move between pages. A node is a page path pattern, such as `/orders/:id`, with the number of flows on it and the share of sessions that reached it (`sessionPercentage`, 0 to 100); its `type` is `entry` when some session started there. A link is a move between two pages by a session's consecutive flows, with how often it happened (`value`) and at most five of the flows it led to (`flowIds`). + +```bash +curl "https://api.keploy.io/client/v1/apps//ui/flow-graph" \ + -H "Authorization: Bearer kep_YOUR_API_KEY" +``` + +```json +{ + "data": { + "nodes": [ + { + "id": "/", + "label": "/", + "type": "entry", + "flowCount": 6, + "sessionPercentage": 100 + }, + { + "id": "/about", + "label": "/about", + "type": "page", + "flowCount": 3, + "sessionPercentage": 50 + }, + { + "id": "/products", + "label": "/products", + "type": "page", + "flowCount": 3, + "sessionPercentage": 50 + }, + { + "id": "/cart", + "label": "/cart", + "type": "page", + "flowCount": 2, + "sessionPercentage": 33.333333333333336 + }, + { + "id": "/checkout", + "label": "/checkout", + "type": "page", + "flowCount": 2, + "sessionPercentage": 33.333333333333336 + } + ], + "links": [ + { + "source": "/", + "target": "/products", + "value": 3, + "flowIds": [ + "c220bcd5-4596-4270-ab4a-accfa4e7a403", + "6368957e-9d8d-405f-939b-c162882cacfc", + "248a8854-06ac-4189-936b-e7eb2bd55326" + ] + }, + { + "source": "/products", + "target": "/cart", + "value": 2, + "flowIds": [ + "453613fc-e400-4461-b184-2321dd2d2559", + "80e82e96-c592-4da3-a3f6-c7f08b5daeec" + ] + }, + { + "source": "/cart", + "target": "/checkout", + "value": 2, + "flowIds": [ + "2635dfaf-b651-4b5d-ab3f-5fb7484f73c3", + "b1e489df-42af-4bd6-b273-fe64147c9a80" + ] + }, + { + "source": "/", + "target": "/about", + "value": 3, + "flowIds": [ + "080dc372-322d-4b09-b644-194917534b1f", + "9195ecd8-dc22-4418-8ccd-23c81aa854d8", + "9888b9ee-f61d-4ca3-b13f-f0524c31d4ed" + ] + } + ], + "totalSessions": 6, + "flowsRead": 16, + "truncated": false, + "backLinksDropped": 0 + }, + "meta": { + "timestamp": "2026-11-20T07:28:16Z" + } +} +``` + +The graph is bounded for an agent's context: the 100 most frequent links, every page they touch, and the most visited other pages up to 200 nodes. `linksDropped` and `nodesDropped`, present when some were left out, count them. Links only go forward in one page order, so moves back to an earlier page are counted in `backLinksDropped`. It is built from the app's 10,000 most recent flows (`truncated` says when there were more). The flow graph is on the REST API and MCP only. + +## Journey plan {#journey-plan} + +`GET /client/v1/apps/{appId}/ui/journeys` · MCP `uiJourneys` · scope `read`, permission `app:view` + +The app's journey plan: how its sessions move between pages, the journeys they take, and the journeys Keploy selects to replay as tests. Every count is of sessions, and every `sessionPercentage` a share of `totalSessions`: a session is one page load of the app, its in-app navigation included, not a person. + +| Query parameter | What it is | +| --------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | +| `refresh` | `true` computes a new plan now, from a minute after the last one was computed; before then the stored plan is answered. It needs `testgen:run` on the app and a key with the `write` scope. | + +```bash +curl "https://api.keploy.io/client/v1/apps//ui/journeys" \ + -H "Authorization: Bearer kep_YOUR_API_KEY" +``` + +```json +{ + "data": { + "planId": "a89602ff-c783-4f8e-bc34-47a151adea19", + "createdAt": 1795159694, + "windowDays": 30, + "from": 1792567694, + "to": 1795159694, + "totalSessions": 6, + "sessionsRead": 6, + "sessionsOnNoPage": 0, + "coveragePermille": 1000, + "coverageTargetPermille": 800, + "maxJourneys": 20, + "fingerprint": "65406ad5a44ff56445ccff79351a05b469be40938f9b8cc917f26f412ac91be7", + "notes": [], + "journeys": [ + { + "id": "2990cf35868e23549307ad2120fe5d58", + "pages": ["/", "/about"], + "sessions": 3, + "sessionPercentage": 50, + "lastSeenAt": 1795159612, + "selected": true, + "pinned": false, + "representativeSessionId": "2a58f279-cc49-48cf-85ab-e7cfd9199a33", + "sdkVersion": "3.1.0", + "flowIds": [ + "d07317d4-e2b5-4ba6-b1db-767966e4e474", + "9888b9ee-f61d-4ca3-b13f-f0524c31d4ed" + ] + }, + { + "id": "778bfd7ec36f1c03bfffa0362a47fe4f", + "pages": ["/", "/products", "/cart", "/checkout"], + "sessions": 2, + "sessionPercentage": 33.333333333333336, + "lastSeenAt": 1795159556, + "selected": true, + "pinned": false, + "representativeSessionId": "55c7ea31-dba1-4d49-acbe-56cf700295e3", + "sdkVersion": "3.1.0", + "flowIds": [ + "f059842e-220d-415d-9195-b909f17f6536", + "c220bcd5-4596-4270-ab4a-accfa4e7a403", + "453613fc-e400-4461-b184-2321dd2d2559", + "2635dfaf-b651-4b5d-ab3f-5fb7484f73c3" + ] + } + ], + "ranked": [ + { + "id": "084e889e39bcbdc4ebbe152a5c81d4ab", + "pages": ["/", "/products"], + "sessions": 3, + "sessionPercentage": 50, + "lastSeenAt": 1795159625, + "selected": false, + "pinned": false, + "representativeSessionId": "db452012-c693-4938-a672-c15f44c59b1a" + } + ], + "rankedDropped": 0, + "nodes": [ + { + "page": "/", + "sessions": 6, + "sessionPercentage": 100, + "entrySessions": 6 + }, + { + "page": "/about", + "sessions": 3, + "sessionPercentage": 50, + "entrySessions": 0 + } + ], + "edges": [ + { + "source": "/", + "target": "/about", + "sessions": 3, + "sessionPercentage": 50 + } + ], + "nodesDropped": 0, + "edgesDropped": 0, + "pagesMerged": 0, + "freshness": { + "computedAt": 1795159694, + "sessionsCovered": 6, + "sessionsProcessedSince": 0, + "recomputesAfterSessions": 1, + "recomputesAt": 1795246094, + "refreshableAt": 1795159754 + } + }, + "meta": { + "timestamp": "2026-11-20T07:28:14Z" + } +} +``` + +The answer above is shortened: `ranked` holds three journeys, `nodes` five pages and `edges` four moves. + +- **`journeys`** are the journeys a test run replays: the most taken ones until they cover `coverageTargetPermille` of the window's moves between pages, at most `maxJourneys`, then any pinned ones. Each has its `pages`, `sessions` and `sessionPercentage`, `lastSeenAt`, and `flowIds`, the recording a test of it replays, which you read with [Get a flow](#get-a-flow). `excluded` says why a journey is not run: `ERRORS_ONLY`, `NO_REPLAYABLE_RECORDING` or `UNSEEN_IN_WINDOW`. +- **`ranked`** are the 50 most taken journeys; **`nodes`** and **`edges`** are the pages and the moves between them. +- **`notes`** say where a bound changed what is counted. +- **`freshness`** says how current the plan is: when it was computed (`computedAt`), how many sessions it read and how many were processed since, after how many more it is recomputed (`recomputesAfterSessions`), when a read recomputes it anyway (`recomputesAt`, a day after `computedAt`), and from when `refresh=true` does (`refreshableAt`). + +A read that computes a plan waits for its turn on the server. When the server's queue of plan computations is full, or the request ends before its plan is ready, the answer is `429` `RATE_LIMITED` with `Retry-After: 10`: retry then. + +## Journey impact {#journey-impact} + +`GET /client/v1/apps/{appId}/ui/journeys/impact` · MCP `uiJourneyImpact` · scope `read`, permission `app:view` + +How many of a plan's sessions took at least one of a set of its journeys. It is a union, never a sum: a session that took two of the journeys counts once. + +| Query parameter | What it is | +| --------------- | ---------------------------------------------------------------------------------------------- | +| `planId` | Required. The plan's `planId`, from the [journey plan](#journey-plan). | +| `journeyIds` | Required. 1 to 100 journey IDs, joined with commas. Each must be one of the plan's `journeys`. | + +```bash +curl "https://api.keploy.io/client/v1/apps//ui/journeys/impact?planId=&journeyIds=," \ + -H "Authorization: Bearer kep_YOUR_API_KEY" +``` + +```json +{ + "data": { + "planId": "a89602ff-c783-4f8e-bc34-47a151adea19", + "journeyIds": [ + "2990cf35868e23549307ad2120fe5d58", + "778bfd7ec36f1c03bfffa0362a47fe4f" + ], + "totalSessions": 6, + "affectedSessions": 5, + "affectedSessionPercentage": 83.33333333333333 + }, + "meta": { + "timestamp": "2026-11-20T07:28:17Z" + } +} +``` + +A journey ID the plan does not run is `400` `INVALID_PARAMETER`; a `planId` that is not one of the app's plans, or a plan older than 14 days, is `404`. + +## List test runs {#list-test-runs} + +`GET /client/v1/apps/{appId}/ui/test-runs` · MCP `listUITestRuns` · scope `read`, permission `app:view` + +The app's UI test runs, newest first, one summary each: `status`, `verdict`, a one-line `summary`, the `planId` it replays, the number of tests, a counter per verdict, the share of sessions its failing journeys stand for (`impact`) and their IDs (`failingJourneyIds`). It takes `offset` and `limit` (at most 100, default 20), and `meta.pagination` says whether more follow. + +A run today comes only from a flow page's **Generate Test** in the dashboard, and it stays `PENDING`: no API call creates or runs one, and the runner that replays journeys as Playwright tests is not in a release build yet. For an app with no runs: + +```bash +curl "https://api.keploy.io/client/v1/apps//ui/test-runs" \ + -H "Authorization: Bearer kep_YOUR_API_KEY" +``` + +```json +{ + "data": [], + "meta": { + "timestamp": "2026-11-20T07:28:18Z", + "pagination": { + "has_next_page": false, + "has_previous_page": false, + "next_cursor": null, + "previous_cursor": null, + "total_count": 0, + "unreadable_count": 0 + } + } +} +``` + +## Get a test run {#get-a-test-run} + +`GET /client/v1/apps/{appId}/ui/test-runs/{runId}` · MCP `getUITestRun` · scope `read`, permission `app:view` + +One run in full: what [List test runs](#list-test-runs) says of it, plus one result per journey (`results`), what to do `next`, and the run's claim by a runner. Each result has the journey's `verdict` (`PASSED`, `FAILED`, `UNCOVERED`, `UNPROVEN`, `NONDETERMINISTIC` or `BLOCKED`), its `cause`, the step it failed at (`failingStep`), and its `sessions` and `sessionPercentage`. A run that is `PENDING` has no results, and its `next` says that no runner has claimed it and not to poll for it. A run of another app is `404`. + +## Delete an app {#delete-an-app} + +`DELETE /client/v1/apps/{appId}` · MCP `deleteApp` · scope `admin` + +```bash +curl -X DELETE "https://api.keploy.io/client/v1/apps/" \ + -H "Authorization: Bearer kep_YOUR_ADMIN_KEY" +``` + +```json +{ + "data": { + "deleted": true + }, + "meta": { + "timestamp": "2026-11-20T07:28:19Z" + } +} +``` + +An OAuth sign-in through the MCP server cannot delete an app: its token never has the `admin` scope. + +## Ingest endpoints {#ingest-endpoints} + +The capture SDK posts what it records to the ingest endpoint, `https://api.keploy.io/ui/v1/ingest`, with the ingest key in `X-API-Key`. These routes take only an ingest key, and any browser origin may call them. + +### Complete a session {#complete-a-session} + +`POST /ui/v1/sessions/{sdkSessionId}/complete`, with the ingest key in `X-API-Key` + +Ends a visit at once, so Keploy turns it into flows within a minute or two instead of after 30 idle minutes, as a test or a CI recording should. `{sdkSessionId}` is the browser's id for the visit, which `KeployCapture.getSessionId()` returns. Close the visit's tab first: a tab left open on a completed session is refused at its next batch. + +```bash +curl -X POST "https://api.keploy.io/ui/v1/sessions//complete" \ + -H "X-API-Key: kep_YOUR_INGEST_KEY" +``` + +```json +{ + "ok": true, + "sessionId": "01056d9a-f9b9-48ef-baf7-15d8804cd7ab", + "status": "COMPLETED" +} +``` + +Completing a session that has already ended answers `200` again, with its current status. A session this app never received is `404`. + +### Answers {#ingest-answers} + +`POST /ui/v1/ingest` is the SDK's: you do not call it yourself. The SDK stops capturing on any `4xx` answer but `429`, and [UI Capture errors and warnings](./errors.md) says what each one means. + +| Status | When | +| ------ | --------------------------------------------------------------------------------------------------------------------- | +| `200` | Accepted. | +| `400` | The request is malformed. | +| `401` | No key, a key that is not a Keploy key or is sent the wrong way, an unknown or expired key, or the key's app is gone. | +| `403` | The key is not an ingest key, such as a personal access token. | +| `404` | No such route under `/ui/v1`, or a complete for a session this app never received. | +| `405` | A method other than `POST`. | +| `409` | A batch for a completed session, a second run of one session id, or a batch id already stored with different events. | +| `413` | A batch over 10 MiB, or a session at its size or batch cap. | +| `415` | A `Content-Type` or `Content-Encoding` the SDK does not send. | +| `429` | The key's rate limit, 6,000 requests a minute, with `Retry-After`. The SDK retries these. | + +## Related + +- [MCP tools](./mcp-tools.md) +- [Operate UI capture from an AI agent](./agents.md) +- [AI agent quickstart](./quickstart-agent.md) +- [Keploy Public API](../running-keploy/public-api.md) +- [UI Capture errors and warnings](./errors.md) diff --git a/versioned_docs/version-4.0.0/ui-capture/sessions-and-sampling.md b/versioned_docs/version-4.0.0/ui-capture/sessions-and-sampling.md index 5a240da36..95f4d2a8a 100644 --- a/versioned_docs/version-4.0.0/ui-capture/sessions-and-sampling.md +++ b/versioned_docs/version-4.0.0/ui-capture/sessions-and-sampling.md @@ -123,7 +123,7 @@ On Keploy's side, a session moves through these states: Extraction runs every minute, and a session's flows are usually ready within a few minutes of its end. A session that cannot be extracted is tried again with growing waits, eight attempts in all over about two hours, and is then marked `FAILED`. A page that was only loaded still gives a flow, of one step: its navigation. -The app's capture status counts its sessions by these states and says what to do next: see [Check that capture is working](../running-keploy/ui-capture.md#check-that-capture-is-working). +The app's capture status counts its sessions by these states and says what to do next: see [Capture status](./rest-api.md#capture-status). ## Related diff --git a/versioned_docs/version-4.0.0/ui-capture/walkthrough-agent.md b/versioned_docs/version-4.0.0/ui-capture/walkthrough-agent.md new file mode 100644 index 000000000..38192a49b --- /dev/null +++ b/versioned_docs/version-4.0.0/ui-capture/walkthrough-agent.md @@ -0,0 +1,671 @@ +--- +id: walkthrough-agent +title: "Walkthrough: an AI agent sets up UI capture and reads flows and journeys" +sidebar_label: An agent onboards a web app +description: Every Keploy MCP tool call an AI agent makes to set up UI capture for a web app and read its flows and journeys, with the real answers. +tags: + - UI testing + - UI capture + - walkthrough + - AI agent + - MCP +keywords: + - AI agent + - Keploy MCP server + - uiCaptureSetup + - uiCaptureStatus + - uiJourneys + - UI flows +--- + +import HowTo from '@site/src/components/HowTo'; +import ProductTier from '@site/src/components/ProductTier'; + +/complete and the ingest key.", url: "#record-visits"}, +{name: "Wait for the flows", text: "Call uiCaptureStatus at most once a minute until it says FLOWS_READY.", url: "#wait-for-the-flows"}, +{name: "Read the flows", text: "Call listUIFlows, then getUIFlow for one flow's steps and network calls.", url: "#read-the-flows"}, +{name: "Read the journeys", text: "Call uiJourneys for the journey plan, and uiJourneyImpact for the share of sessions a set of its journeys stands for.", url: "#read-the-journeys"}, +{name: "Clean up", text: "Revoke the ingest key with revokeUIIngestKey, and delete the app with deleteApp and an admin key.", url: "#clean-up"}, +]} +/> + + + +This walkthrough follows an AI agent through UI capture from start to finish, through the Keploy MCP server: it finds the tools, creates an app, mints an ingest key, adds the capture SDK to the [React sample shop](https://github.com/keploy/samples-typescript/tree/main/ui-capture-react-shop), checks that visits arrive, and reads the flows, the journey plan and its shares of sessions. Then it cleans up. + +Each step shows the tool call as the agent's MCP client sends it to `invoke_tool`, its answer, and what the agent does next. Every answer is from a real run with six visits to the shop, shortened and without its `meta` (but for the test runs' pagination), with IDs replaced and the server written as `api.keploy.io`. To connect your own agent first, follow the [AI agent quickstart](./quickstart-agent.md). + +:::note Beta +UI Testing is in beta. Recording visits and reading them as flows and journeys works today. Replaying journeys as Playwright tests is in beta too, and its runner is not in a release build yet, so the agent here reads journeys and test runs and does not run tests. See [What an agent cannot do yet](#what-an-agent-cannot-do-yet). +::: + +## What the agent does {#what-the-agent-does} + +1. Finds the UI capture tools. +2. Creates an app. +3. Mints an ingest key. +4. Adds the capture SDK to the shop. +5. Checks that capture is waiting for visits. +6. Records visits and ends each one. +7. Waits for the flows. +8. Reads the flows. +9. Reads the journey plan. +10. Counts the share of sessions a set of journeys stands for. +11. Reads the test runs. +12. Cleans up. + +## 1. Find the tools {#find-the-tools} + +The server runs in tool-search mode: its `tools/list` holds only `search_tools`, `get_tool_schema`, `invoke_tool` and four sign-in and setup tools. Its instructions, in the answer to the MCP `initialize` request, have a UI CAPTURE section that names the UI capture tools in the order an agent uses them. So the agent fetches all their schemas in one call: + +```json title="get_tool_schema" +{ + "names": [ + "createApp", + "listApps", + "getApp", + "deleteApp", + "uiCaptureSetup", + "listUIIngestKeys", + "revokeUIIngestKey", + "uiCaptureStatus", + "listUIFlows", + "getUIFlow", + "uiFlowGraph", + "uiJourneys", + "uiJourneyImpact", + "listUITestRuns", + "getUITestRun" + ] +} +``` + +The answer has each tool's description and input schema. Every UI capture tool takes `appId`; [MCP tools](./mcp-tools.md#tools) lists the rest of their inputs. From here on, the agent runs each tool with `invoke_tool`, with the tool's `name` and `arguments`. + +## 2. Create the app {#create-the-app} + +The agent creates an app for the shop. An app for UI capture needs only a name, and the name cannot be changed later. + +```json title="invoke_tool" +{ + "name": "createApp", + "arguments": { + "name": "docs-demo-agent" + } +} +``` + +```json title="Answer (shortened)" +{ + "data": { + "createdAt": 1795159503, + "id": "af1ec8ee-c43a-4c9c-826b-4223f506fb19", + "name": "docs-demo-agent" + }, + "status_code": 201 +} +``` + +A tool's answer is the REST API's answer, with the HTTP status added as `status_code`. The app's `id` is the `appId` every other tool takes. + +## 3. Mint an ingest key {#mint-an-ingest-key} + +```json title="invoke_tool" +{ + "name": "uiCaptureSetup", + "arguments": { + "appId": "af1ec8ee-c43a-4c9c-826b-4223f506fb19", + "name": "storefront web", + "ttlDays": 30 + } +} +``` + +```json title="Answer" +{ + "data": { + "apiKey": "kep_YOUR_INGEST_KEY", + "endpoint": "https://api.keploy.io/ui/v1/ingest", + "key": { + "appId": "af1ec8ee-c43a-4c9c-826b-4223f506fb19", + "createdAt": 1795159503, + "expiresAt": 1797751503, + "id": "ec6ebf01-0250-49c5-b71c-554cfec6f0af", + "keyPrefix": "kep_YOUR_ING", + "name": "storefront web", + "status": "active" + }, + "next": [ + "apiKey is shown only in this response: store it now (it is safe to ship in the page; it can only post recordings into this app)", + "install the SDK and call KeployCapture.init once at app start-up, in the browser bundle", + "browse the app, then GET /client/v1/apps/af1ec8ee-c43a-4c9c-826b-4223f506fb19/ui/capture-status (MCP: uiCaptureStatus) to see sessions arrive", + "a session becomes flows 30 minutes after its last activity (to end one at once, as a test or CI recording does, POST https://api.keploy.io/ui/v1/sessions/{sdkSessionId}/complete with this key in X-API-Key); read them with GET /client/v1/apps/af1ec8ee-c43a-4c9c-826b-4223f506fb19/ui/flows (MCP: listUIFlows)", + "once sessions are processed, GET /client/v1/apps/af1ec8ee-c43a-4c9c-826b-4223f506fb19/ui/journeys (MCP: uiJourneys) is the journey plan a UI test run replays; GET /client/v1/apps/af1ec8ee-c43a-4c9c-826b-4223f506fb19/ui/test-runs (MCP: listUITestRuns, getUITestRun) reads the runs" + ], + "sdk": { + "init": "import { KeployCapture } from \"@keploy/capture-sdk\";\n\nawait KeployCapture.init({\n endpoint: \"https://api.keploy.io/ui/v1/ingest\",\n apiKey: \"kep_YOUR_INGEST_KEY\",\n});\n", + "install": "npm install @keploy/capture-sdk", + "package": "@keploy/capture-sdk" + } + }, + "status_code": 201 +} +``` + +`apiKey` appears only in this answer, so the agent keeps it now. It is safe to ship in the shop's JavaScript: it can only post recordings into this one app. `ttlDays` is optional; without it the key does not expire. Each call mints a new key. + +## 4. Add the SDK {#add-the-sdk} + +The answer's `sdk.install` and `sdk.init` are what the agent adds to a frontend: install `@keploy/capture-sdk`, and call `KeployCapture.init` once, in a module the app imports before it makes any request. The sample shop already does both: `src/main.tsx` imports `src/keploy.ts` first, and `src/keploy.ts` calls `init` with the endpoint and key from `.env`. So the agent only writes `.env`: + +```bash title=".env" +VITE_KEPLOY_INGEST_ENDPOINT=https://api.keploy.io/ui/v1/ingest +VITE_KEPLOY_INGEST_KEY=kep_YOUR_INGEST_KEY +``` + +Then it builds the shop and serves the build: `npm run build && npm run preview`. + +The `init` snippet sets only `endpoint` and `apiKey`, so the SDK runs with its default `privacyMode: "balanced"`, which masks everything visitors type. The shop writes the default out. For a public site, the agent would set `"strict"`, which masks all visible text too. + +## 5. Check that capture is waiting for visits {#check-capture} + +```json title="invoke_tool" +{ + "name": "uiCaptureStatus", + "arguments": { + "appId": "af1ec8ee-c43a-4c9c-826b-4223f506fb19" + } +} +``` + +```json title="Answer (shortened)" +{ + "data": { + "explanation": "no session from the capture SDK has reached this app", + "idleAfterSeconds": 1800, + "state": "NO_SESSIONS", + "totalFlows": 0, + "totalSessions": 0 + }, + "status_code": 200 +} +``` + +`NO_SESSIONS` until someone uses the app. The answer's `next` lists what to check if nothing arrives. + +## 6. Record visits {#record-visits} + +Here six visits were made in a headless Chromium, along three paths through the shop. A person using the shop by hand records the same way. + +| Visits | Pages | What each does on the way | +| ------- | ----------------------------------------- | ----------------------------------------------------------- | +| 1, 2 | `/` → `/products` → `/cart` → `/checkout` | Searches for `kettle`, adds it to the cart, places an order | +| 3, 4, 5 | `/` → `/about` | | +| 6 | `/` → `/products` | Searches for `kettle` | + +While the first visit was still open, `uiCaptureStatus` said it was recording: + +```json title="Answer to uiCaptureStatus (shortened)" +{ + "data": { + "explanation": "sessions are arriving; a visit becomes flows 30 minutes after its last activity", + "idleAfterSeconds": 1800, + "lastBatchAt": 1795159548, + "next": [ + "keep using the app, or wait 30 minutes after the last activity, then check again", + "to end a visit at once (a test or CI recording), POST https://api.keploy.io/ui/v1/sessions/{sdkSessionId}/complete with the app's ingest key in X-API-Key (the SDK's own session id for the visit); its flows follow within a minute or two" + ], + "recentSessions": [ + { + "batches": 5, + "eventCount": 70, + "id": "a1e271a4-c6e8-456a-ab38-83f4bf64db4f", + "lastBatchAt": 1795159548, + "sdkSessionId": "01056d9a-f9b9-48ef-baf7-15d8804cd7ab", + "sdkVersion": "3.1.0", + "startedAt": 1795159526, + "status": "RECORDING" + } + ], + "sessionsByStatus": { + "RECORDING": 1 + }, + "state": "RECORDING", + "totalFlows": 0, + "totalSessions": 1 + }, + "status_code": 200 +} +``` + +A visit becomes flows `idleAfterSeconds`, 30 minutes, after its last activity. To have its flows at once, the agent ends each visit: it closes the visit's tab, then completes the session with the ingest key. `sdkSessionId` is the browser's own id for the visit: `uiCaptureStatus` lists it under `recentSessions`, as above, and it is what `KeployCapture.getSessionId()` returns, which the shop prints in the browser console: + +```bash +curl -X POST "https://api.keploy.io/ui/v1/sessions//complete" \ + -H "X-API-Key: kep_YOUR_INGEST_KEY" +``` + +```json title="Answer" +{ + "ok": true, + "sessionId": "01056d9a-f9b9-48ef-baf7-15d8804cd7ab", + "status": "COMPLETED" +} +``` + +A tab left open on a completed session is refused at its next batch, which is why the agent closes it first. + +## 7. Wait for the flows {#wait-for-the-flows} + +The agent calls `uiCaptureStatus` once a minute. Flows are extracted every minute, so polling faster gains nothing. + +```json title="Answer to uiCaptureStatus (shortened)" +{ + "data": { + "explanation": "flows have been extracted from recorded sessions", + "next": [ + "read them with GET /client/v1/apps/af1ec8ee-c43a-4c9c-826b-4223f506fb19/ui/flows (MCP: listUIFlows)", + "each flow lists its steps and the network calls it made" + ], + "recentSessions": [ + { + "batches": 3, + "completedAt": 1795159633, + "eventCount": 26, + "id": "db452012-c693-4938-a672-c15f44c59b1a", + "lastBatchAt": 1795159632, + "sdkSessionId": "a1a5cfc9-a48a-483f-8139-26e7ad540e8b", + "sdkVersion": "3.1.0", + "startedAt": 1795159625, + "status": "PROCESSED" + } + ], + "sessionsByStatus": { + "PROCESSED": 6 + }, + "state": "FLOWS_READY", + "totalFlows": 16, + "totalSessions": 6 + }, + "status_code": 200 +} +``` + +`FLOWS_READY` as soon as one flow exists, even while other visits are still being processed: a minute before this answer, it already said `FLOWS_READY`, with 4 sessions `PROCESSED` and 2 still `COMPLETED`. To know that every visit is in, the agent checks `sessionsByStatus`. + +## 8. Read the flows {#read-the-flows} + +```json title="invoke_tool" +{ + "name": "listUIFlows", + "arguments": { + "appId": "af1ec8ee-c43a-4c9c-826b-4223f506fb19", + "limit": 50 + } +} +``` + +```json title="Answer (shortened to 2 of 16 flows)" +{ + "data": [ + { + "clusterId": "661be522-1276-4612-9e7e-181dfd1ad3ff", + "createdAt": 1795159660, + "description": "2 steps, 1 API calls, 6.1s", + "id": "d07317d4-e2b5-4ba6-b1db-767966e4e474", + "name": "Click on /", + "networkCalls": 1, + "sessionId": "2a58f279-cc49-48cf-85ab-e7cfd9199a33", + "status": "CLUSTERED", + "steps": 2 + }, + { + "clusterId": "4949ebdb-3873-4f10-9d79-c2027ee534a8", + "createdAt": 1795159660, + "description": "1 steps, 1 API calls, 0.0s", + "id": "9888b9ee-f61d-4ca3-b13f-f0524c31d4ed", + "name": "Navigate on /about", + "networkCalls": 1, + "sessionId": "2a58f279-cc49-48cf-85ab-e7cfd9199a33", + "status": "CLUSTERED", + "steps": 1 + } + ], + "status_code": 200 +} +``` + +The six visits made 16 flows, one for each page each visit opened. A flow's name is a verb and the page it happened on, or `Verb: /a → /b` for a flow that ends on another page: `Fill` when something was typed, else `Click` when something was clicked, else `Navigate`. The agent reads one flow's steps and calls with `getUIFlow`: + +```json title="invoke_tool" +{ + "name": "getUIFlow", + "arguments": { + "appId": "af1ec8ee-c43a-4c9c-826b-4223f506fb19", + "flowId": "248a8854-06ac-4189-936b-e7eb2bd55326" + } +} +``` + +```json title="Answer" +{ + "data": { + "clusterId": "16ba6a5d-a195-4ebb-9457-7f697bfc55c7", + "createdAt": 1795159660, + "description": "3 steps, 2 API calls, 6.1s", + "id": "248a8854-06ac-4189-936b-e7eb2bd55326", + "name": "Fill on /products", + "networkCallList": [ + { + "method": "GET", + "responseStatus": 200, + "timestamp": 1795159625571, + "url": "http://localhost:4173/api/products" + }, + { + "method": "GET", + "responseStatus": 200, + "timestamp": 1795159631678, + "url": "http://localhost:4173/api/products" + } + ], + "networkCalls": 2, + "outcome": { + "elements": [], + "viewport": { + "height": 720, + "width": 1280 + } + }, + "sessionId": "db452012-c693-4938-a672-c15f44c59b1a", + "status": "CLUSTERED", + "stepList": [ + { + "action": "navigate", + "hasValue": true, + "index": 0, + "pageUrl": "http://localhost:4173/products", + "selector": "", + "timestamp": 1795159625548 + }, + { + "action": "input", + "hasValue": true, + "index": 1, + "pageUrl": "http://localhost:4173/products", + "selector": "[data-testid=\"search-input\"]", + "timestamp": 1795159631613 + }, + { + "action": "click", + "index": 2, + "pageUrl": "http://localhost:4173/products", + "selector": "[data-testid=\"search-button\"]", + "timestamp": 1795159631661 + } + ], + "steps": 3 + }, + "status_code": 200 +} +``` + +This is visit 6 on the Products page. A step says that a value was typed (`hasValue`), never what. The first call is the page loading its products; the second is the search, `GET /api/products?q=kettle`, which comes back without its query. [What an agent sees](./agents.md#what-an-agent-sees) lists what is left out. Step and call `timestamp`s are in milliseconds. + +## 9. Read the journey plan {#read-the-journeys} + +```json title="invoke_tool" +{ + "name": "uiJourneys", + "arguments": { + "appId": "af1ec8ee-c43a-4c9c-826b-4223f506fb19" + } +} +``` + +```json title="Answer (shortened)" +{ + "data": { + "coveragePermille": 1000, + "coverageTargetPermille": 800, + "freshness": { + "computedAt": 1795159694, + "recomputesAfterSessions": 1, + "recomputesAt": 1795246094, + "refreshableAt": 1795159754, + "sessionsCovered": 6, + "sessionsProcessedSince": 0 + }, + "journeys": [ + { + "flowIds": [ + "d07317d4-e2b5-4ba6-b1db-767966e4e474", + "9888b9ee-f61d-4ca3-b13f-f0524c31d4ed" + ], + "id": "2990cf35868e23549307ad2120fe5d58", + "lastSeenAt": 1795159612, + "pages": ["/", "/about"], + "pinned": false, + "representativeSessionId": "2a58f279-cc49-48cf-85ab-e7cfd9199a33", + "sdkVersion": "3.1.0", + "selected": true, + "sessionPercentage": 50, + "sessions": 3 + }, + { + "flowIds": [ + "f059842e-220d-415d-9195-b909f17f6536", + "c220bcd5-4596-4270-ab4a-accfa4e7a403", + "453613fc-e400-4461-b184-2321dd2d2559", + "2635dfaf-b651-4b5d-ab3f-5fb7484f73c3" + ], + "id": "778bfd7ec36f1c03bfffa0362a47fe4f", + "lastSeenAt": 1795159556, + "pages": ["/", "/products", "/cart", "/checkout"], + "pinned": false, + "representativeSessionId": "55c7ea31-dba1-4d49-acbe-56cf700295e3", + "sdkVersion": "3.1.0", + "selected": true, + "sessionPercentage": 33.333333333333336, + "sessions": 2 + } + ], + "maxJourneys": 20, + "planId": "a89602ff-c783-4f8e-bc34-47a151adea19", + "ranked": [ + { + "id": "084e889e39bcbdc4ebbe152a5c81d4ab", + "pages": ["/", "/products"], + "selected": false, + "sessionPercentage": 50, + "sessions": 3 + }, + { + "id": "2990cf35868e23549307ad2120fe5d58", + "pages": ["/", "/about"], + "selected": true, + "sessionPercentage": 50, + "sessions": 3 + }, + { + "id": "778bfd7ec36f1c03bfffa0362a47fe4f", + "pages": ["/", "/products", "/cart", "/checkout"], + "selected": true, + "sessionPercentage": 33.333333333333336, + "sessions": 2 + } + ], + "totalSessions": 6, + "windowDays": 30 + }, + "status_code": 200 +} +``` + +- **`journeys`** are the journeys Keploy selected to replay as tests: `/` → `/about`, taken by 3 of the 6 sessions, and `/` → `/products` → `/cart` → `/checkout`, taken by 2. Together they cover all the moves the sessions made between pages (`coveragePermille` 1000), past the 800 Keploy aims for. Each one's `flowIds` are the recording a test of it replays. +- **`ranked`** are the most taken journeys. `/` → `/products` was taken by 3 sessions but is not selected: the longer journey that continues it was, and its test replays those pages too. +- **`freshness`** says how current the plan is. `recomputesAfterSessions: 1` means the next processed session makes a read compute a new plan. `refresh: true` computes one from `refreshableAt`, a minute after `computedAt`, and needs `testgen:run`. + +## 10. Count the share of sessions {#share-of-sessions} + +`uiJourneyImpact` counts the sessions that took at least one of a set of the plan's journeys, as the share of sessions a failure in those journeys would stand for. `journeyIds` is one string, the IDs joined with commas: + +```json title="invoke_tool" +{ + "name": "uiJourneyImpact", + "arguments": { + "appId": "af1ec8ee-c43a-4c9c-826b-4223f506fb19", + "planId": "a89602ff-c783-4f8e-bc34-47a151adea19", + "journeyIds": "2990cf35868e23549307ad2120fe5d58,778bfd7ec36f1c03bfffa0362a47fe4f" + } +} +``` + +```json title="Answer" +{ + "data": { + "affectedSessionPercentage": 83.33333333333333, + "affectedSessions": 5, + "journeyIds": [ + "2990cf35868e23549307ad2120fe5d58", + "778bfd7ec36f1c03bfffa0362a47fe4f" + ], + "planId": "a89602ff-c783-4f8e-bc34-47a151adea19", + "totalSessions": 6 + }, + "status_code": 200 +} +``` + +5 of the 6 sessions, 83.3%: a union, never a sum. Each ID must be one of the plan's `journeys`. A journey that is only in `ranked` is refused: + +```json title="invoke_tool" +{ + "name": "uiJourneyImpact", + "arguments": { + "appId": "af1ec8ee-c43a-4c9c-826b-4223f506fb19", + "planId": "a89602ff-c783-4f8e-bc34-47a151adea19", + "journeyIds": "084e889e39bcbdc4ebbe152a5c81d4ab" + } +} +``` + +```json title="Answer" +{ + "error": { + "code": "INVALID_PARAMETER", + "message": "invalid UI journey: journey is not in the plan: 084e889e39bcbdc4ebbe152a5c81d4ab" + }, + "status_code": 400 +} +``` + +## 11. Read the test runs {#read-test-runs} + +```json title="invoke_tool" +{ + "name": "listUITestRuns", + "arguments": { + "appId": "af1ec8ee-c43a-4c9c-826b-4223f506fb19" + } +} +``` + +```json title="Answer" +{ + "data": [], + "meta": { + "pagination": { + "has_next_page": false, + "has_previous_page": false, + "next_cursor": null, + "previous_cursor": null, + "total_count": 0, + "unreadable_count": 0 + }, + "timestamp": "2026-11-20T07:28:17Z" + }, + "status_code": 200 +} +``` + +The app has no runs. No tool creates a run: a run today comes only from a flow page's **Generate Test** in the dashboard, and it stays `PENDING`, because the runner that replays journeys as Playwright tests is not in a release build yet. + +## 12. Clean up {#clean-up} + +The agent revokes the ingest key, which it lists with `listUIIngestKeys` when it does not have the `keyId`: + +```json title="invoke_tool" +{ + "name": "revokeUIIngestKey", + "arguments": { + "appId": "af1ec8ee-c43a-4c9c-826b-4223f506fb19", + "keyId": "ec6ebf01-0250-49c5-b71c-554cfec6f0af" + } +} +``` + +```json title="Answer" +{ + "data": { + "id": "ec6ebf01-0250-49c5-b71c-554cfec6f0af", + "revoked": true + }, + "status_code": 200 +} +``` + +Pages that still use the key are refused within about 30 seconds. Then it deletes the app, which needs a key with the `admin` scope. An OAuth sign-in cannot delete apps. + +```json title="invoke_tool" +{ + "name": "deleteApp", + "arguments": { + "appId": "af1ec8ee-c43a-4c9c-826b-4223f506fb19" + } +} +``` + +```json title="Answer" +{ + "data": { + "deleted": true + }, + "status_code": 200 +} +``` + +## Waits {#waits} + +- **A visit becomes flows** 30 minutes after its last activity, or a minute or two after `POST /ui/v1/sessions//complete`. +- **Poll `uiCaptureStatus` at most once a minute.** No state lasts forever: a visit that cannot be processed is retried for about two hours, then counted as failed. +- **The journey plan:** a read computes a new one once enough sessions have been processed since the last. `refresh: true` forces one at most once a minute. A `429` `RATE_LIMITED` answer to `uiJourneys` means the server's queue of plan computations is full, or the plan was not ready in time: its message says to retry in 10 seconds. +- **A `PENDING` test run** is not claimed by any runner: do not poll for it. + +[Waiting and polling](./agents.md#waiting-and-polling) has the details. + +## What an agent cannot do yet {#what-an-agent-cannot-do-yet} + +- Generate or run tests: no tool or REST route creates, claims or completes a UI test run, and the runner, `keploy ui run`, is not in a release build yet. +- Pin a journey into the plan, or accept a change. +- Run UI tests in CI: no service-account role runs them. A CI job can check capture status and read the plan with a key bound to the `UI_READER` role, on a paid plan. + +## Related + +- [AI agent quickstart](./quickstart-agent.md) +- [Operate UI capture from an AI agent](./agents.md) +- [MCP tools](./mcp-tools.md) +- [REST API](./rest-api.md) +- [From first visit to journeys](./walkthrough-dashboard.md) diff --git a/versioned_docs/version-4.0.0/ui-capture/walkthrough-dashboard.md b/versioned_docs/version-4.0.0/ui-capture/walkthrough-dashboard.md index 980f06d2a..86493c0c4 100644 --- a/versioned_docs/version-4.0.0/ui-capture/walkthrough-dashboard.md +++ b/versioned_docs/version-4.0.0/ui-capture/walkthrough-dashboard.md @@ -299,7 +299,7 @@ curl https://api.keploy.io/client/v1/apps//ui/flows/ \ } ``` -This is the flow from step 5, shortened to two of its five steps and two of its four calls. A flow's step and call `timestamp`s are in milliseconds. The API returns less than the dashboard: a step says that a value was typed (`hasValue`) and never what, and URLs come without their query, so the search's call has no `?q=kettle`. [What an agent sees](../running-keploy/ui-capture.md#what-an-agent-sees) lists what is left out. +This is the flow from step 5, shortened to two of its five steps and two of its four calls. A flow's step and call `timestamp`s are in milliseconds. The API returns less than the dashboard: a step says that a value was typed (`hasValue`) and never what, and URLs come without their query, so the search's call has no `?q=kettle`. [What an agent sees](./agents.md#what-an-agent-sees) lists what is left out. **Journeys.** The same plan as the **Journeys** tab: @@ -341,7 +341,7 @@ curl https://api.keploy.io/client/v1/apps//ui/journeys \ This response is shortened too. It also carries `ranked`, the most taken journeys, with `/` → `/products` first; `nodes` and `edges`, the graph's pages and moves; and `freshness`, which says when the plan was computed. `flowIds` are the recording a test of the journey replays, each a flow you can read with `GET …/ui/flows/`. -The Keploy MCP server has the same calls as tools: `uiCaptureStatus`, `listUIFlows`, `getUIFlow` and `uiJourneys`. See [From an AI agent (MCP)](../running-keploy/ui-capture.md#set-up-ui-capture-from-an-ai-agent-mcp). +The Keploy MCP server has the same calls as tools: `uiCaptureStatus`, `listUIFlows`, `getUIFlow` and `uiJourneys`. See [MCP tools](./mcp-tools.md). ## 8. Clean up {#clean-up} @@ -361,7 +361,7 @@ curl -X DELETE https://api.keploy.io/client/v1/apps/ \ ## What's next {#whats-next} - [Install the SDK in your own app](../running-keploy/ui-capture.md#install-the-sdk), as early in its startup as you can. -- [Set up UI capture from an AI agent](../running-keploy/ui-capture.md#set-up-ui-capture-from-an-ai-agent-mcp) through the Keploy MCP server. +- [Set up UI capture from an AI agent](./quickstart-agent.md) through the Keploy MCP server. - [UI Capture errors and warnings](./errors.md): what each `KUI_E_` code means and how to fix it. ## Related diff --git a/versioned_sidebars/version-4.0.0-sidebars.json b/versioned_sidebars/version-4.0.0-sidebars.json index 12eaeb757..8b779635e 100644 --- a/versioned_sidebars/version-4.0.0-sidebars.json +++ b/versioned_sidebars/version-4.0.0-sidebars.json @@ -294,14 +294,20 @@ "label": "Quickstarts", "collapsible": true, "collapsed": false, - "items": ["ui-capture/quickstart-dashboard"] + "items": [ + "ui-capture/quickstart-dashboard", + "ui-capture/quickstart-agent" + ] }, { "type": "category", "label": "Walkthroughs", "collapsible": true, "collapsed": false, - "items": ["ui-capture/walkthrough-dashboard"] + "items": [ + "ui-capture/walkthrough-dashboard", + "ui-capture/walkthrough-agent" + ] }, { "type": "category", @@ -312,7 +318,8 @@ "ui-capture/privacy", "ui-capture/sessions-and-sampling", "ui-capture/network-and-bodies", - "ui-capture/ingest-keys" + "ui-capture/ingest-keys", + "ui-capture/agents" ] }, { @@ -331,6 +338,9 @@ "id": "ui-capture/refusals", "label": "Refused verbs" }, + "ui-capture/rest-api", + "ui-capture/mcp-tools", + "ui-capture/cli", { "type": "doc", "id": "ui-capture/limits",