diff --git a/plugins/flowstudio-power-automate/README.md b/plugins/flowstudio-power-automate/README.md index 7178f7fd1d..387c49e452 100644 --- a/plugins/flowstudio-power-automate/README.md +++ b/plugins/flowstudio-power-automate/README.md @@ -1,6 +1,6 @@ # FlowStudio Power Automate Plugin -Give your AI agent the same visibility you have in the Power Automate portal. The Graph API only returns top-level run status — agents can't see action inputs, loop iterations, nested failures, or who owns a flow. Flow Studio MCP exposes all of it. +Give your AI agent the same visibility you have in the Power Automate portal. The Graph API only returns top-level run status — agents can't see action inputs, loop iterations, nested failures, or who owns a flow. FlowStudio MCP exposes all of it. This plugin includes five skills covering the full lifecycle: connect, debug, build, monitor, and govern Power Automate cloud flows. @@ -16,7 +16,7 @@ Requires a [FlowStudio MCP](https://mcp.flowstudio.app) subscription. | Flow health and failure rates | Nothing | | Who built a flow, what connectors it uses | Nothing | -Flow Studio MCP fills these gaps. +FlowStudio MCP fills these gaps. ## Installation diff --git a/skills/flowstudio-power-automate-build/SKILL.md b/skills/flowstudio-power-automate-build/SKILL.md index 8723722980..25c05dca97 100644 --- a/skills/flowstudio-power-automate-build/SKILL.md +++ b/skills/flowstudio-power-automate-build/SKILL.md @@ -411,50 +411,43 @@ print(f"Status: {result['responseStatus']}, via: {result['invocation']}") print(result.get("warning")) # set when a required input was missing: the run still ran, with null ``` -### Brand-new non-HTTP flows (Recurrence, connector triggers, etc.) +### Brand-new non-HTTP flows -A brand-new Recurrence or connector-triggered flow has **no prior runs** to -resubmit and no HTTP endpoint to call. This is the ONLY scenario where you -need the temporary HTTP trigger approach below. **Deploy with a temporary -HTTP trigger first, test the actions, then swap to the production trigger.** +A brand-new **Recurrence** flow needs no workaround: deploy it, then run it +immediately with `trigger_live_flow` and no `body` — same as the portal's +"Run flow" button. A body is refused; scheduled triggers take no inputs. -Compact recipe: +A brand-new **connector-triggered** flow (SharePoint, webhooks) has no prior +runs and cannot fire without a real source event. Deploy with a temporary +HTTP trigger, test the actions, then swap to the production trigger: ```python production_trigger = definition["triggers"] definition["triggers"] = { "manual": {"type": "Request", "kind": "Http", "inputs": {"schema": {}}} } - -result = mcp("update_live_flow", - environmentName=ENV, +result = mcp("update_live_flow", environmentName=ENV, flowName=FLOW_ID, # omit if creating new - definition=definition, - connectionReferences=connection_references, + definition=definition, connectionReferences=connection_references, displayName="Overdue Invoice Notifications") -FLOW_ID = FLOW_ID or result["created"] +FLOW_ID = FLOW_ID or result["flowName"] -test = mcp("trigger_live_flow", environmentName=ENV, flowName=FLOW_ID, - body={"sample": "payload"}) +mcp("trigger_live_flow", environmentName=ENV, flowName=FLOW_ID, + body={"sample": "payload"}) runs = mcp("get_live_flow_runs", environmentName=ENV, flowName=FLOW_ID, top=1) - if runs[0]["status"] == "Failed": err = mcp("get_live_flow_run_error", environmentName=ENV, flowName=FLOW_ID, runName=runs[0]["name"]) raise Exception(err["failedActions"][-1]) definition["triggers"] = production_trigger -mcp("update_live_flow", - environmentName=ENV, - flowName=FLOW_ID, - definition=definition, - connectionReferences=connection_references) +mcp("update_live_flow", environmentName=ENV, flowName=FLOW_ID, + definition=definition, connectionReferences=connection_references) ``` The trigger is only the entry point; testing through HTTP still exercises the same actions. If actions use `triggerBody()` or `triggerOutputs()`, pass a -representative `trigger_live_flow.body` shaped like the production trigger -payload. +representative `body` shaped like the production trigger payload. --- diff --git a/skills/flowstudio-power-automate-debug/SKILL.md b/skills/flowstudio-power-automate-debug/SKILL.md index c4d1ddd9df..50622de97f 100644 --- a/skills/flowstudio-power-automate-debug/SKILL.md +++ b/skills/flowstudio-power-automate-debug/SKILL.md @@ -409,10 +409,10 @@ print(new_runs[0]["status"]) # Succeeded = done | Scenario | Use | Why | |---|---|---| | **Testing a fix** on any flow | `resubmit_live_flow_run` | Replays the exact trigger payload that caused the failure — best way to verify | -| Recurrence / scheduled flow | `resubmit_live_flow_run` | Cannot be triggered on demand any other way | +| Recurrence / scheduled flow | `trigger_live_flow` (no `body`) | Runs it now, like the portal's "Run flow" button; resubmit replays a past run's data | | SharePoint / connector trigger | `resubmit_live_flow_run` | Cannot be triggered without creating a real SP item | | HTTP, Button, or PowerApps trigger with **custom** test payload | `trigger_live_flow` | When you need to send different data than the original run | -| Brand-new flow, never run | `trigger_live_flow` (HTTP, Button, PowerApps) | No prior run exists to resubmit | +| Brand-new flow, never run | `trigger_live_flow` | No prior run exists to resubmit | ### Testing HTTP, Button, and PowerApps flows with custom payloads @@ -445,8 +445,10 @@ if result.get("warning"): ``` > `trigger_live_flow` handles AAD-authenticated triggers automatically. -> Works for `Request` triggers only: HTTP request, Button, and PowerApps. -> Scheduled and connector triggers cannot be run this way. +> Works for `Request` triggers (HTTP request, Button, PowerApps) and for +> scheduled (Recurrence) flows, which it runs immediately — with no `body`, +> since a scheduled trigger takes no inputs (a body is refused). Automated +> connector triggers only fire from their source event. > > Power Automate does not enforce a trigger's `required` inputs. If you leave > one out the run still starts, with that input null, and the result carries a diff --git a/skills/flowstudio-power-automate-governance/SKILL.md b/skills/flowstudio-power-automate-governance/SKILL.md index 58092bcc14..b4413ea6d7 100644 --- a/skills/flowstudio-power-automate-governance/SKILL.md +++ b/skills/flowstudio-power-automate-governance/SKILL.md @@ -64,18 +64,18 @@ If a deleted flow has `monitor=true`, suggest disabling monitoring ## The Write Tool: `update_store_flow` -`update_store_flow` writes governance metadata to the **Flow Studio cache +`update_store_flow` writes governance metadata to the **FlowStudio cache only** — it does NOT modify the flow in Power Automate. These fields are not visible via `get_live_flow` or the PA portal. They exist only in the -Flow Studio store and are used by Flow Studio's scanning pipeline and +FlowStudio store and are used by FlowStudio's scanning pipeline and notification rules. This means: -- `ownerTeam` / `supportEmail` — sets who Flow Studio considers the +- `ownerTeam` / `supportEmail` — sets who FlowStudio considers the governance contact. Does NOT change the actual PA flow owner. -- `rule_notify_email` — sets who receives Flow Studio failure/missing-run +- `rule_notify_email` — sets who receives FlowStudio failure/missing-run notifications. Does NOT change Microsoft's built-in flow failure alerts. -- `monitor` / `critical` / `businessImpact` — Flow Studio classification +- `monitor` / `critical` / `businessImpact` — FlowStudio classification only. Power Automate has no equivalent fields. Merge semantics — only fields you provide are updated. Returns the full @@ -218,7 +218,7 @@ store tags, so read/append/write. Avoid overriding computed `tier` unless asked. ### 7. Maker Offboarding When an employee leaves, identify their flows and apps, and reassign -Flow Studio governance contacts and notification recipients. +FlowStudio governance contacts and notification recipients. ``` 1. get_store_maker(makerKey="") @@ -234,7 +234,7 @@ Flow Studio governance contacts and notification recipients. apps needing manual reassignment ``` -This changes Flow Studio governance contacts, not actual PA ownership. Power +This changes FlowStudio governance contacts, not actual PA ownership. Power Apps ownership changes are manual/admin-center work. ### 8. Security Review diff --git a/skills/flowstudio-power-automate-mcp/SKILL.md b/skills/flowstudio-power-automate-mcp/SKILL.md index de1d9bd485..517390c7c7 100644 --- a/skills/flowstudio-power-automate-mcp/SKILL.md +++ b/skills/flowstudio-power-automate-mcp/SKILL.md @@ -26,6 +26,9 @@ skills that all build on this one. > compatible Power Automate MCP server). You will need: > - MCP endpoint: `https://mcp.flowstudio.app/mcp` (same for all subscribers) > - API key / JWT token (`x-api-key` header — NOT Bearer) +> - In ChatGPT or claude.ai there is no key: add `https://mcp.flowstudio.app/mcp/oauth` +> as a connector and sign in with Microsoft — see the +> [ChatGPT walkthrough](https://learn.flowstudio.app/chatgpt-power-automate) > - Power Platform environment name (e.g. `Default-`) --- diff --git a/skills/flowstudio-power-automate-mcp/references/tool-reference.md b/skills/flowstudio-power-automate-mcp/references/tool-reference.md index 060eea6187..efd4f7f49e 100644 --- a/skills/flowstudio-power-automate-mcp/references/tool-reference.md +++ b/skills/flowstudio-power-automate-mcp/references/tool-reference.md @@ -93,9 +93,9 @@ Response: wrapper object with `connections` array. > Filter by status: prefer `overallStatus == "Connected"` when present; otherwise > check `statuses[0].status == "Connected"`. > -> For build workflows, pass `environmentName` to avoid using a connection from -> the wrong environment. Omit it only when intentionally inventorying connections -> across all environments. +> `environmentName` is required — the platform cannot list connections across +> environments (omitting it answers 400 MissingEnvironmentFilter). Get one from +> `list_live_environments`. > > Pass `search=` to narrow output and receive > `connectionReferenceTemplate` plus `hostTemplate` values that can be copied @@ -448,9 +448,10 @@ Response keys: `flowKey`, `triggerName`, `triggerKind`, `invocation`, `triggerUr `requiresAadAuth`, `authType`, `responseStatus`, `responseBody`, `runName`, and `warning` when a required trigger input was not supplied. -> **Works for `Request` triggers: HTTP request, Button, and PowerApps.** Returns an -> error for Recurrence and connector triggers: -> `"only HTTP Request triggers can be invoked via this tool"`. +> **Works for `Request` triggers (HTTP request, Button, PowerApps) and for +> scheduled (Recurrence) flows.** A scheduled flow runs immediately, like the +> portal's "Run flow" button, and takes no `body` — one is refused. Automated +> connector triggers only fire from their source event and return an error. > > HTTP triggers go through the signed callback URL (`invocation: callbackUrl`). > Button and PowerApps triggers have no callback URL; with a `body` they run @@ -483,8 +484,7 @@ Response keys: `flowKey`, `triggerName`, `triggerKind`, `invocation`, `triggerUr ### `set_live_flow_state` -Start or stop a Power Automate flow via the live PA API. Does **not** require -a Power Clarity workspace — works for any flow the impersonated account can access. +Start or stop a Power Automate flow in any environment you have access to. Reads the current state first and only issues the start/stop call if a change is actually needed. @@ -594,11 +594,11 @@ tool schemas cannot tell you. connectionReferences. Use `set_live_flow_state` to start/stop a flow. ### `trigger_live_flow` -- **Works for HTTP, Button, and PowerApps triggers.** Returns error for Recurrence, - connector, and other trigger types. +- **Works for HTTP, Button, PowerApps, and scheduled (Recurrence) triggers.** + A scheduled flow runs with no `body`. Automated connector triggers error. - Pass trigger inputs as `body`. A `warning` in the result means a required input was missing and the run started with it null. -- AAD-authenticated triggers are handled automatically (impersonated Bearer token). +- AAD-authenticated triggers are handled automatically (Bearer token attached). ### `get_live_flow_runs` - `top` defaults to **30** with automatic pagination for higher values. @@ -611,8 +611,9 @@ tool schemas cannot tell you. - `poster`: `"Flow bot"` for Workflows bot identity, `"User"` for user identity. ### `list_live_connections` -- For build workflows, pass `environmentName`; omitting it inventories - connections across environments. +- `environmentName` is required; the platform cannot list connections across + environments. `top` is applied after `search` and the result carries + `truncated` + `matchedCount` when the cap drops matches. - Use `search=` to get smaller output and paste-ready `connectionReferenceTemplate` / `hostTemplate` values. - `id` is the value you need for `connectionName` in `connectionReferences`. diff --git a/skills/flowstudio-power-automate-monitoring/SKILL.md b/skills/flowstudio-power-automate-monitoring/SKILL.md index bbd1a24867..cdf7457d03 100644 --- a/skills/flowstudio-power-automate-monitoring/SKILL.md +++ b/skills/flowstudio-power-automate-monitoring/SKILL.md @@ -39,7 +39,7 @@ enriched with governance metadata and remediation hints. ## How Monitoring Works -Flow Studio scans the Power Automate API daily for each subscriber and caches +FlowStudio scans the Power Automate API daily for each subscriber and caches the results. There are two levels: - **All flows** get metadata scanned: definition, connections, owners, trigger @@ -54,7 +54,7 @@ the results. There are two levels: a flow was last scanned. If stale, the scanning pipeline may not be running. **Enabling monitoring:** Set `monitor: true` via `update_store_flow` or the -Flow Studio for Teams app +FlowStudio for Teams app ([how to select flows](https://learn.flowstudio.app/teams-monitoring)). **Designating critical flows:** Use `update_store_flow` with `critical=true`