implement-leaf-ticket
Skill0 starsChanged 7 days agoMITView on GitHub
Implement the next leaf GitHub issue end to end: build, review, commit, push, close.
In this repository14 other files it gives its agents.
- docxSkill
- pdfSkill
- pptxSkill
- xlsxSkill
- code-review-seniorSkill
- commit-message-generatorSkill
- explain-in-vietnamesesSkill
- security-auditSkill
- ponytail-auditSkill
- ponytail-debtSkill
- ponytail-gainSkill
- ponytail-helpSkill
- ponytail-reviewSkill
- ponytailSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
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 10
Review: keep fixing and re-running "code-review" until a pass comes back with no findings.
≈ “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
- Pick the next leaf: an open issue with no open sub-issues and no open blockers. If a ticket is passed, search only its sub-issue tree (it is the leaf if it has none); otherwise search all open issues. Take the first in its parent's order (else lowest number). Skip leaves only a human can do.
- Build it: call the Skill tool with "ponytail", then read and follow
.agents/skills/implement/SKILL.md(user-only, so the Skill tool can't load it). - Clean up where needed: remove throwaway work created for this task and code made redundant by the implementation. Keep cleanup within the task's scope and run the relevant tests on the final result.
- Reconcile related material before closing: update affected docs in the same change, and update the ticket and related parent/blocker issues to reflect the delivered scope, acceptance checklist, and any remaining work. Keep the issue open if its acceptance criteria are still unmet.
- Review: keep fixing and re-running "code-review" until a pass comes back with no findings.
- Ship: once cleanup, related updates, tests, and review are complete, commit and push (child repos before the parent's submodule pin), then close the issue with a comment naming the commits.
Throughout:
- Steps written for a human are yours whenever you can do or verify them; the owner's approval is standing.
- Delegate to subagents (they may spawn their own), use cheaper models for small tasks. Tell code-writing subagents to load "ponytail".
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.