Skip to content
DevOps AI ToolKit
Newsletter
All prompts
AI for Slack Difficulty: Intermediate ClaudeChatGPTCursor

Slack Stars API Personal Ops Triage Queue

Build a personal follow-up queue for on-call engineers using the Slack stars API (saved items) so starred alerts, threads, and files become a triageable to-do list that clears itself as work completes.

Target user
On-call and platform engineers who lose track of alerts and follow-ups scattered across Slack channels
Difficulty
Intermediate
Tools
Claude, ChatGPT, Cursor

The prompt

You are a senior SRE who lives in Slack during on-call and has built lightweight tooling on top of the Slack stars ("saved items") API so nothing important falls through the cracks between alert firing and follow-up landing.

I will provide:
- The kinds of items engineers need to track (alert messages, incident threads, log-file uploads, follow-up asks)
- The channels/apps those items originate from
- How the team currently loses track of them today

Your job:

1. **Model the queue** — treat a starred item as a triage-queue entry. Explain the relevant stars methods (`stars.add`, `stars.remove`, `stars.list`) and the shape of a star: it references a `channel` + `message` (ts), a `file`, or a `file_comment`. Note that stars are per-user (a personal queue), and call out that the stars API is on Slack's deprecation/limited path so the design must abstract the storage behind an interface (stars today, bookmarks/datastore/DB tomorrow) — do not hard-couple.

2. **Auto-enqueue on signal** — wire an Events API / reaction handler so a chosen emoji (e.g. :eyes: or :inbox_tray:) on any message calls `stars.add` for the acting user, turning "I'll deal with this later" into a tracked item. Show the handler.

3. **List & render the queue** — call `stars.list` (paginated with `cursor`), then render a Block Kit summary (App Home tab or scheduled DM) grouping items by age, with a deep link (`chat.getPermalink`) to each source message and a "Done" button per item.

4. **Clear on completion** — the "Done" button calls `stars.remove` for that item and updates the view. Reacting with a completion emoji should also dequeue. Show both paths.

5. **Staleness & escalation** — flag items older than a threshold (e.g. 24h) and optionally DM a nudge or, for on-call, roll unresolved items into the handoff summary. Show the staleness computation from the star's `date_create`.

6. **Pagination & rate limits** — `stars.list` is a Tier 3 method; show cursor pagination and backoff on `ratelimited` so a large queue does not throttle.

7. **Migration seam** — because the stars API is deprecated for some token types, provide a `TriageStore` interface (add/remove/list) with a stars-backed implementation and a stub DB-backed one, so swapping the backend later is a one-file change.

Output as: (a) the reaction-driven `stars.add`/`stars.remove` handlers, (b) the paginated `stars.list` fetch with backoff, (c) the App Home Block Kit render with permalinks and Done buttons, (d) the staleness/escalation logic, (e) the `TriageStore` interface with two implementations.

Bias toward: abstracting storage behind an interface given the stars API's uncertain future, per-user personal queues (not shared state), and always deep-linking back to the source so triage is one click from context.

Run this prompt with AI

Test it, get an AI-improved version, or compare models — live in the Prompt Workspace. No copy-paste.

Related prompts

More Slack prompts & error guides

Browse every Slack prompt and troubleshooting guide in one place.

Free download · 368-page PDF

Reading prompts? Get all 500 in one free PDF

500 battle-tested, copy-paste AI prompts engineered by a senior systems engineer — every one with fill-in placeholders and safety/back-out notes. Drop your email and it's yours.

  • 500 prompts: Linux · Kubernetes · Terraform · OpenStack · GitLab · Docker · Monitoring · Incident Response
  • Instant PDF download — yours free, forever
  • Plus one practical AI-workflow email a week (no spam)

Single opt-in · unsubscribe anytime · no spam.