finalize-loop
Skill2 starsChanged 9 days agoMITView on GitHub
Use when completed implementation needs final review, verification, an authorized commit, or a handoff; not for every intermediate edit.
In this repository15 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- capability-incident-diagnoserSkill
- context-managementSkill
- goal-workflowSkill
- host-cli-usageSkill
- tool-useSkill
- fast-checksSkill
- filesystem-wizardSkill
- lean-codeSkill
- test-freshnessSkill
- security-boundary-reviewSkill
- portable-shellSkill
- process-debuggingSkill
- brokered-git-operatorSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 14
Do not impose a fixed ceremony or rerun unchanged checks just to finish.
- Distinctive: line 16
Before landing, establish that the request is satisfied, required checks pass, and the intended diff is reviewed.
- Distinctive: line 22
Commit, push, or open a review request only within applicable authority.
- Distinctive: line 26
Read back the authoritative state after the final mutation: for example, working-tree status and the resulting commit, remote ref, or …
≈ “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.
Files
Close the evidence gaps
Compare the complete change set with the request, including staged and untracked files. Look for unrelated edits, debug leftovers, sensitive data, and generated artifacts. Preserve work that is not yours.
Reuse current verification results; run the owning gate and any missing focused coverage. Review and checks may interleave as findings require. Do not impose a fixed ceremony or rerun unchanged checks just to finish.
Before landing, establish that the request is satisfied, required checks pass, and the intended diff is reviewed. Report baseline failures separately; an explained failure is not permission to bypass a required gate. Repair regressions within scope and the agreed budget. At a blocker or limit, preserve the work and report the missing evidence.
Commit, push, or open a review request only within applicable authority. Existing authorization need not be requested again. If publication is not authorized, leave the result local and say so.
Read back the authoritative state after the final mutation: for example, working-tree status and the resulting commit, remote ref, or review request. A final report should identify that state, checks and results, and remaining limits. The evidence must support completion; it need not be the last tool call or force the report into a transcript format.
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.

