gc
Skill6 starsChanged 49 days agoMITView on GitHub
Git catch-up — bring a stale local checkout in sync with the remote before trusting it. Use before claiming code is missing/unmerged, before starting a ticket, or when a report looks wrong: fetch, compare against origin, and reconcile.
In this repository24 other files it gives its agents.
- docs/AGENTS.mdAGENTS.md
- atomic-reportSkill
- board-validateSkill
- data-modelSkill
- delivery-hygieneSkill
- dev-reportSkill
- dev-setupSkill
- frontend-qualitySkill
- git-branchSkill
- land-and-archiveSkill
- orchestration-healthSkill
- orchestratorSkill
- pipeline-qualitySkill
- plan-updateSkill
- project-planSkill
- release-gateSkill
- repo-auditSkill
- scaleSkill
- security-reviewSkill
- swarmSkill
- test-coverageSkill
- tpmSkill
- trackerSkill
- worktree-cleanupSkill
The file
Files
gc — Git Catch-Up
Local checkouts drift behind the remote. Never claim something is missing, unmerged, or stale from a checkout you haven't synced. Fetch first, then judge.
The routine
- Fetch everything:
git fetch --all --prune - See where you stand vs. the remote default branch:
git status -sb # ahead/behind for the current branch git log --oneline origin/main -5 # what's actually on the remote - Reconcile:
- Behind, clean tree →
git pull --ff-only(or rebase your branch ontoorigin/main). - Local changes → stash or commit to your ticket branch first, then update.
- "Missing" code → check
origin/mainand other branches before concluding it's gone; it usually landed on a branch your checkout hadn't fetched.
- Behind, clean tree →
- Prune what's dead: deleted-on-remote branches are cleaned by
--prune; drop merged local branches (seeworktree-cleanup).
Why this is a discipline, not a nicety
A checkout that's a few commits behind will confidently report already-merged work as
"unmerged" and already-fixed bugs as "still broken." Audits, status reports, and "is this
done?" answers must be made against the remote, not against a stale local main. When in
doubt: git fetch before you assert.
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
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.

