Data
Weekly product digest from Slack, Linear and GitHub
Combine user feedback, shipped work, and open Linear items into a weekly digest that a human reviews before it posts.
slacklineargithubdigestweeklyproductreportingchangelogdataautomation
[ workflow / data ]
Weekly product digest from Slack, Linear and GitHub
Each week, Cosmos reviews feedback channels, merged PRs, closed issues, and Linear boards from the past seven days. It turns those signals into themes, shipped work, blockers, and a short narrative summary. A human curator reviews the draft before it posts so the digest has judgment, not just aggregation.
08 nodes
07 edges
Trigger[trigger]
Weekly schedule
Monday 09:00 / manual
System step[slack]
Collect Slack signals
Feedback channels, past 7 days
System step[github]
Collect GitHub activity
Merged PRs + closed issues
System step[linear]
Collect Linear activity
Shipped + blocked tickets
AI Agent step[synthesise]
Synthesise digest
Themes, shipping, narrative
Output / Result[post-draft]
Post draft to curator channel
Slack blocks + approve button
Human-in-the-loop[review]
Curator reviews draft
Approve / edit / discard
Output / Result[publish]
Publish to team channel
Final digest posted
Workflow prompt
Paste this into Augment to reproduce the workflow end-to-end.
Build a Cosmos workflow that produces a curated weekly product digest from Slack, GitHub, and Linear. Trigger: weekly schedule (e.g. every Monday at 09:00 in the team's timezone), or manually triggered. Steps: 1. Collect Slack signals. Search the configured feedback channels (e.g. #feedback-*, #bugs, #customer-success) for messages from the past 7 days. Exclude bot messages and emoji-only reactions. Extract: the top themes by frequency, verbatim quotes for the most impactful messages, and any threads marked with a bookmark or specific emoji. 2. Collect GitHub activity. Pull merged pull requests and closed issues from the past 7 days across all configured repositories. Group by: new features, bug fixes, refactors, dependency updates, and infra changes. Extract PR titles, linked issue descriptions, and authors. 3. Collect Linear activity. Pull tickets that moved to "Done" or "Shipped" in the past 7 days. Pull tickets that are "Blocked" or "In Review" and have been in that state for more than a configurable threshold (default: 3 days). Group by project. 4. Synthesise the digest. Produce a structured document with the following sections: - **This week in shipping**: a bulleted list of shipped features and notable fixes, sourced from GitHub + Linear. - **What users are saying**: the top 3–5 feedback themes with representative quotes from Slack. - **Still in flight**: open work that is blocked or stalled, with context on the blocker. - **Summary**: a short (3–5 sentence) narrative paragraph that captures the week's direction and the most important signal. 5. Post the draft digest as a Slack message to a designated curator channel (e.g. #digest-review), formatted as blocks for easy reading. Include an "Approve and post" button and an "Edit" link. 6. Human curator reviews the draft. They can approve-and-publish, edit the draft inline, or discard. 7. Once approved, post the final digest to the main product or engineering channel. Constraints: - Always include the curator review step: never publish a digest without a human reading it first. - Use verbatim Slack quotes sparingly and only when they add color that a paraphrase would lose. - Keep the digest under 600 words so it is actually read.