From 2c1fc8f95c6bb234a7a72fd8ac3b22fd357b544e Mon Sep 17 00:00:00 2001 From: biswaroop1547 Date: Fri, 10 Jul 2026 04:01:30 +0530 Subject: [PATCH 1/4] docs: publish customer release notes MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Source releases: https://github.com/premAI-io/premapp-backend/releases/tag/prod-v1.3.0 https://github.com/premAI-io/fluso-frontend/releases/tag/prod-v1.3.0 Hand-assembled following the release-notes skill (SKILL.md + release-notes-playbook.md) after the publish-release-notes workflow's Claude step failed authentication (401) on two consecutive live dispatches — CLAUDE_CODE_OAUTH_TOKEN in this repo needs rotating. Co-Authored-By: Claude Fable 5 --- release-notes.mdx | 40 ++++++++++++++++++++++++++++++++++++++++ 1 file changed, 40 insertions(+) diff --git a/release-notes.mdx b/release-notes.mdx index 0781038..b27dc7d 100644 --- a/release-notes.mdx +++ b/release-notes.mdx @@ -3,6 +3,46 @@ title: "Release notes" description: "What's new in Fluso — product updates, improvements, and fixes." --- + + +**Highlights** +- Steer a running reply, or queue what's next, and nothing is lost to a refresh, a dropped connection, or a stop. +- Projects now remember where they stood — pick one back up after a while and ask "where did we leave off," and you get a real answer instead of re-explaining yourself. +- Chat streams live: replies and tool activity now appear as they happen instead of arriving all at once after a delay. +- Confidential Mode is now live-verified. The Encrypted badge opens a view that cryptographically re-checks the confidential server in real time, and the plan is renamed to Confidential. +- Tasks that use your connected apps (GitHub, Linear, Slack, and more) finish in about half the time, with far fewer wrong-tool guesses. +- A hung or unresponsive assistant now recovers in a couple of minutes with nothing lost, instead of a wait that could previously stretch past half an hour. + +**Polish & experience** +- Running a task from a task card now gives the assistant the full task details right away, not just its title. +- In-app voice messages show their real length instead of a stuck 0:00. +- A Markdown file in the file viewer shows one Download button instead of two. + +**Connectors & sync** +- Multi-step tasks that use your connected apps (GitHub, Linear, Notion, Slack, and the rest) finish in about half the time, with far fewer wrong-tool guesses or repeated attempts. +- The assistant reaches for the right app-specific guide before it acts on a connector instead of guessing, and asks you to connect an account it isn't already connected to. + +**Assistant reliability** +- Send a message while the assistant is working and it either steers the current reply or joins an editable queue to run next, with edit, send-now, and cancel controls. None of it depends on keeping the tab open: refresh, close it, or come back on another device later, and it is exactly where you left it. +- A dropped connection partway through a long file or command now retries and resumes on its own instead of failing the turn. +- The action chips under a task no longer get permanently stuck if one model hiccups — it falls back automatically, with a Retry button for a genuine network blip. +- Reloading mid-reply, or leaving a tab open a while, can no longer leave a conversation showing your message with no answer, or blank it entirely. +- Stopping a turn now reads "Stopped after Xs" instead of the misleading "Worked for Xs". + +**Smarter assistant** +- Projects carry a running sense of their own state (current work, open items, files touched) into every message, on top of your existing preferences. Teach it a convention once (a house rule, a style, a correction) and it holds until you change it, without leaking into a different project. +- Reading a Word document no longer leaks hidden formatting junk, and documents the assistant creates for you now get real Word tables instead of literal Markdown pipe syntax. + +**Onboarding & profile** +- Bring your Claude or ChatGPT conversation history in — a new import flow on the Projects page turns an exported zip into a real thread in your workspace. +- An expired sign-in code can now be resent from the verification screen instead of starting over. + +**Reliability, privacy & trust** +- Confidential Mode is now live-verified: the Encrypted badge opens a view that cryptographically re-checks the confidential server in real time, covering the attested CPU and GPU and the model being served. The plan is also renamed to Confidential for sponsored accounts. +- A focused internal security audit closed six findings this cycle, spanning token validation, account isolation, and file-handling safety. + + + **Highlights** From b8eef0c20bf471ef6bbfd4bba7d298b4524ccdad Mon Sep 17 00:00:00 2001 From: biswaroop1547 Date: Fri, 10 Jul 2026 04:13:37 +0530 Subject: [PATCH 2/4] docs: revise Desktop 1.3.0 highlights per review feedback - Lead with import (Claude/ChatGPT/Codex/any agent app, full context) and project working memory instead of steer/queue and streaming - Drop the confidential-mode and steer/queue Highlights lines (their fuller detail already lives in their own themed sections) - Explain context pruning in plain terms: fewer tool calls, less token budget used - Rephrase working memory around what it means and why, not just what changed - Cut a connector bullet clause describing already-existing behavior - Reframe the docx line around the read/create capability, not a bug fix - Trim near-duplicate phrasing between Highlights and themed sections Co-Authored-By: Claude Fable 5 --- release-notes.mdx | 16 +++++++--------- 1 file changed, 7 insertions(+), 9 deletions(-) diff --git a/release-notes.mdx b/release-notes.mdx index b27dc7d..2954013 100644 --- a/release-notes.mdx +++ b/release-notes.mdx @@ -6,10 +6,8 @@ description: "What's new in Fluso — product updates, improvements, and fixes." **Highlights** -- Steer a running reply, or queue what's next, and nothing is lost to a refresh, a dropped connection, or a stop. -- Projects now remember where they stood — pick one back up after a while and ask "where did we leave off," and you get a real answer instead of re-explaining yourself. -- Chat streams live: replies and tool activity now appear as they happen instead of arriving all at once after a delay. -- Confidential Mode is now live-verified. The Encrypted badge opens a view that cryptographically re-checks the confidential server in real time, and the plan is renamed to Confidential. +- Switching from ChatGPT, Codex, or another AI app? Bring the whole conversation with you — import turns an export into a real, continuous thread here, so you're not starting from zero. +- Projects now have working memory — Fluso keeps a running sense of what's done, what's open, and how you like things handled, so picking one back up never feels like starting over. - Tasks that use your connected apps (GitHub, Linear, Slack, and more) finish in about half the time, with far fewer wrong-tool guesses. - A hung or unresponsive assistant now recovers in a couple of minutes with nothing lost, instead of a wait that could previously stretch past half an hour. @@ -19,8 +17,8 @@ description: "What's new in Fluso — product updates, improvements, and fixes." - A Markdown file in the file viewer shows one Download button instead of two. **Connectors & sync** -- Multi-step tasks that use your connected apps (GitHub, Linear, Notion, Slack, and the rest) finish in about half the time, with far fewer wrong-tool guesses or repeated attempts. -- The assistant reaches for the right app-specific guide before it acts on a connector instead of guessing, and asks you to connect an account it isn't already connected to. +- The assistant is lighter on its feet now: old tool results get cleared out once they're no longer needed, instead of being carried forward turn after turn, so a task takes fewer tool calls and uses less of its token budget along the way. +- The assistant reaches for the right app-specific guide before it acts on a connector, instead of guessing. **Assistant reliability** - Send a message while the assistant is working and it either steers the current reply or joins an editable queue to run next, with edit, send-now, and cancel controls. None of it depends on keeping the tab open: refresh, close it, or come back on another device later, and it is exactly where you left it. @@ -30,11 +28,11 @@ description: "What's new in Fluso — product updates, improvements, and fixes." - Stopping a turn now reads "Stopped after Xs" instead of the misleading "Worked for Xs". **Smarter assistant** -- Projects carry a running sense of their own state (current work, open items, files touched) into every message, on top of your existing preferences. Teach it a convention once (a house rule, a style, a correction) and it holds until you change it, without leaking into a different project. -- Reading a Word document no longer leaks hidden formatting junk, and documents the assistant creates for you now get real Word tables instead of literal Markdown pipe syntax. +- Working memory means each project remembers itself: open items, recent files, and any convention you've taught it (a house rule, a style, a correction) carry forward automatically, on top of your regular preferences, and stay scoped to that project. Come back after a few weeks and ask "where did we leave off" and you'll get a real answer, not a blank page. +- The assistant reads and creates Word documents more reliably now: no hidden formatting junk when it reads one, and real Word tables, not raw Markdown, in the documents it creates for you. **Onboarding & profile** -- Bring your Claude or ChatGPT conversation history in — a new import flow on the Projects page turns an exported zip into a real thread in your workspace. +- A new import flow lives on the Projects page — upload an exported chat from Claude, ChatGPT, Codex, or another agent app, and it becomes a real thread here, ready to continue. - An expired sign-in code can now be resent from the verification screen instead of starting over. **Reliability, privacy & trust** From b072b6647e584a06a6542833431f5af131de60c4 Mon Sep 17 00:00:00 2001 From: biswaroop1547 Date: Fri, 10 Jul 2026 04:20:35 +0530 Subject: [PATCH 3/4] docs: restore steer/queue highlight, add doc links, fix import framing MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit - Restore steer & follow-up queue as a Highlight (reordered after import/memory, not removed — it's still a major change) - Correct the import bullets to match features/imports.mdx: it creates real projects with context/decisions/files, not a continued chat thread, which is what "context migration, not chat migration" means - Link docs.fluso.ai/features/{imports,memory,confidential} inline at the three points where a real, matching doc page exists Co-Authored-By: Claude Fable 5 --- release-notes.mdx | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/release-notes.mdx b/release-notes.mdx index 2954013..21c1ebe 100644 --- a/release-notes.mdx +++ b/release-notes.mdx @@ -6,8 +6,9 @@ description: "What's new in Fluso — product updates, improvements, and fixes." **Highlights** -- Switching from ChatGPT, Codex, or another AI app? Bring the whole conversation with you — import turns an export into a real, continuous thread here, so you're not starting from zero. -- Projects now have working memory — Fluso keeps a running sense of what's done, what's open, and how you like things handled, so picking one back up never feels like starting over. +- Bring your context in from ChatGPT, Claude, or another agent app (Codex and more) — [import](https://docs.fluso.ai/features/imports) turns an old export into real Fluso projects, with the decisions, notes, and files intact, so you're not starting from zero. +- Projects now have [working memory](https://docs.fluso.ai/features/memory) — Fluso keeps a running sense of what's done, what's open, and how you like things handled, so picking one back up never feels like starting over. +- Steer a running reply, or queue what's next — nothing is lost to a refresh, a dropped connection, or a stop. - Tasks that use your connected apps (GitHub, Linear, Slack, and more) finish in about half the time, with far fewer wrong-tool guesses. - A hung or unresponsive assistant now recovers in a couple of minutes with nothing lost, instead of a wait that could previously stretch past half an hour. @@ -32,11 +33,11 @@ description: "What's new in Fluso — product updates, improvements, and fixes." - The assistant reads and creates Word documents more reliably now: no hidden formatting junk when it reads one, and real Word tables, not raw Markdown, in the documents it creates for you. **Onboarding & profile** -- A new import flow lives on the Projects page — upload an exported chat from Claude, ChatGPT, Codex, or another agent app, and it becomes a real thread here, ready to continue. +- The import flow lives on the Projects page: upload the export and Fluso sorts it into projects automatically, with a running thread you can watch while it works and a summary when it's done. - An expired sign-in code can now be resent from the verification screen instead of starting over. **Reliability, privacy & trust** -- Confidential Mode is now live-verified: the Encrypted badge opens a view that cryptographically re-checks the confidential server in real time, covering the attested CPU and GPU and the model being served. The plan is also renamed to Confidential for sponsored accounts. +- [Confidential Mode](https://docs.fluso.ai/features/confidential) is now live-verified: the Encrypted badge opens a view that cryptographically re-checks the confidential server in real time, covering the attested CPU and GPU and the model being served. The plan is also renamed to Confidential for sponsored accounts. - A focused internal security audit closed six findings this cycle, spanning token validation, account isolation, and file-handling safety. From 6d23e1df3582528eaadd2951f8db8ae19f23aba9 Mon Sep 17 00:00:00 2001 From: biswaroop1547 Date: Fri, 10 Jul 2026 04:23:50 +0530 Subject: [PATCH 4/4] docs: expand steer/queue into a proper hours-or-days durability story MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Ties the durable server-side queue together with the reliability work around it (wedge recovery, connection retry, reload/blank fixes) into one coherent reason you can queue up work, walk away, and trust it's still there — rather than leaving it as a single one-line bullet. Avoids repeating the wedge-recovery Highlights bullet verbatim. Co-Authored-By: Claude Fable 5 --- release-notes.mdx | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/release-notes.mdx b/release-notes.mdx index 21c1ebe..af457f5 100644 --- a/release-notes.mdx +++ b/release-notes.mdx @@ -8,7 +8,7 @@ description: "What's new in Fluso — product updates, improvements, and fixes." **Highlights** - Bring your context in from ChatGPT, Claude, or another agent app (Codex and more) — [import](https://docs.fluso.ai/features/imports) turns an old export into real Fluso projects, with the decisions, notes, and files intact, so you're not starting from zero. - Projects now have [working memory](https://docs.fluso.ai/features/memory) — Fluso keeps a running sense of what's done, what's open, and how you like things handled, so picking one back up never feels like starting over. -- Steer a running reply, or queue what's next — nothing is lost to a refresh, a dropped connection, or a stop. +- Steer a running reply, or queue up what's next — durable enough to walk away for hours, or a day, and find it exactly as you left it. - Tasks that use your connected apps (GitHub, Linear, Slack, and more) finish in about half the time, with far fewer wrong-tool guesses. - A hung or unresponsive assistant now recovers in a couple of minutes with nothing lost, instead of a wait that could previously stretch past half an hour. @@ -22,7 +22,7 @@ description: "What's new in Fluso — product updates, improvements, and fixes." - The assistant reaches for the right app-specific guide before it acts on a connector, instead of guessing. **Assistant reliability** -- Send a message while the assistant is working and it either steers the current reply or joins an editable queue to run next, with edit, send-now, and cancel controls. None of it depends on keeping the tab open: refresh, close it, or come back on another device later, and it is exactly where you left it. +- Send a message while the assistant is working and it either steers the current reply or joins an editable queue to run next, with edit, send-now, and cancel controls. The queue lives on the server, not your browser tab, so you can line up a stretch of work, close the laptop, and come back an hour later, or the next day, to find it exactly where you left it — steer and queue are built for real, unattended runs, not just quick back-and-forth, backed by the reliability work below. - A dropped connection partway through a long file or command now retries and resumes on its own instead of failing the turn. - The action chips under a task no longer get permanently stuck if one model hiccups — it falls back automatically, with a Retry button for a genuine network blip. - Reloading mid-reply, or leaving a tab open a while, can no longer leave a conversation showing your message with no answer, or blank it entirely.