docs: release notes — Desktop 1.3.0#30
Merged
Merged
Conversation
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 <noreply@anthropic.com>
- 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 <noreply@anthropic.com>
- 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 <noreply@anthropic.com>
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 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Customer-facing release notes generated from:
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
Published 2026-07-09. Leads with importing history from Claude, ChatGPT, Codex, or any other agent app (full conversation, not a summary) and project-level working memory, then faster/more-accurate connector use (with plain-language context-pruning framing: fewer tool calls, less token budget), and assistant-reliability fixes (wedged-workspace recovery, reload/blanking races, task-chip fallback). Steer/follow-up queue and Confidential Mode's live verification are covered in their own themed sections rather than Highlights. Security hardening is folded into Reliability, privacy & trust, calm and non-specific per house style. Backend-only work (CI, infra, deploy) has no user-facing effect and is omitted.
Note on how this PR was produced: the
publish-release-notesworkflow's Claude step failed authentication (401 Invalid authentication credentials) on two consecutive live dispatches — confirmed to beCLAUDE_CODE_OAUTH_TOKENitself being rejected by Anthropic's API, not a GitHub/OIDC permissions issue (that step succeeds fine). This entry was hand-assembled following the samerelease-notesskill (SKILL.md+release-notes-playbook.md) the workflow uses, then revised per direct review feedback, and validated locally withcspellandmint broken-links(both clean). The secret needs rotating for future automated runs.Review the rendered diff and merge to publish.
mainrequires an approving review, so your review is the gate — merging deploys to docs.fluso.ai.🤖 Generated with Claude Code