agentleFS
Sign inStart free
loganware05loganware05/captains-compass-cursor/.cursor/skills/northstar-connected-routine/SKILL.md

northstar-connected-routine

Skill2 starsChanged 33 days agoMITView on GitHub

Runs the NorthStar connected operating routine across Slack, Linear, GitHub, and Cursor with fail-closed Captain approval

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    After approval: issue, rollback, branch/worktree, Linear children, signed work packet, dispatch only the configured Cursor agent to the …

    line 38
  2. Very specific:

    After REVIEW_READY, --propose-roles stages persistent-role drafts only (Captain PR required).

    line 60
  3. Distinctive:

    No automatic merge, release, live Skill install, or destructive production action

    line 68

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

NorthStar Connected Routine

Use this Skill when

Coordinating an objective across Slack intake, Linear ledger, GitHub engineering truth, and Cursor execution under the NorthStar M21 connected operating model.

Product identity

  • Canonical name: NorthStar (former name: Captain's Compass)
  • Governance terms Captain and First Mate are unchanged
  • Legacy aliases resolve to the same NorthStar project/run via orchestrator/branding.py

Prerequisites

  • Approved IMPLEMENTATION_PLAN.md before any Cursor dispatch
  • Fixture adapters for CI (mode: fixtures) — no live credentials in tests
  • Live mode (--mode live) requires env secrets + sandbox product allowlist
  • M21 integration agent id: bc-05d4594d-fac7-4378-b595-c20e3c006044

Procedure

  1. Intake from allowlisted Slack (@NorthStar), Linear NorthStar project, GitHub issue with northstar label, GitHub webhook ingress, or explicit Captain Cursor instruction.
  2. Normalize event; reject replay/cross-project; create run_id.
  3. Reconcile context; propose plan; stop at AWAITING_CAPTAIN_APPROVAL.
  4. Canonical approval = GitHub Captain decision + matching plan digest (NORTHSTAR_APPROVE plan_digest=<64-hex>). Slack/Linear may record intent only. Live mode refuses the CLI --approve shortcut.
  5. After approval: issue, rollback, branch/worktree, Linear children, signed work packet, dispatch only the configured Cursor agent to the sandbox product allowlist.
  6. Publish Slack transitions only: work received, plan awaiting approval, execution started, blocked/budget stopped, review ready, completed.
  7. Validate, open/update PR, stop at AWAITING_MERGE for Captain merge.

CLI

./scripts/run-northstar-routine.sh --demo
./scripts/run-northstar-routine.sh --demo --approve --advance-to-review
./scripts/run-northstar-routine.sh --mode live --product-repo loganware05/captain-compass-sandbox \
  --provider github --event path/to/event.json
./scripts/serve-northstar-ingress.sh --mode live --bind 127.0.0.1 --port 8787
./scripts/run-northstar-routine.sh --demo --approve --advance-to-review \
  --propose-roles --surface-routing --notion-mode fixtures
./scripts/reconcile-northstar-run.sh --run .agent/evidence/<run_id>/run.json

See docs/integrations/northstar-live-ops.md for env vars and fail-closed rules.

After REVIEW_READY, --propose-roles stages persistent-role drafts only (Captain PR required). --surface-routing lists pending routing proposals and does not apply weights. Use ./scripts/apply-routing-proposal.sh with captain_approved for bounded apply. --notion-mode fixtures|live gathers non-authoritative research context / summary mirrors.

Safety

  • No automatic merge, release, live Skill install, or destructive production action
  • Missing GitHub stops the routine; missing Slack suppresses notifications; missing Linear falls back to GitHub issues; missing Cursor leaves a launch-ready work packet
  • Secrets never enter prompts, logs, Slack, Linear, GitHub, evidence, or fixtures
  • Product dispatch allowlist: loganware05/captain-compass-sandbox only

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free