agentleFS
lucastosettolucastosetto/dev-skills/skills/implement/SKILL.md

implement

Skill0 starsChanged 25 days agoMITView on GitHub

Implement a ticket end to end in an isolated git worktree: read requirements, spec, QA plan, code, verify, document, commit locally, then queue push, staging deploy and PR for confirmation. Works in any repo with any tracker. Use when the user gives a ticket key/URL or requirements and says "implement", "build this", "fix this bug", "work on this ticket", "pick this up", "/implement".

191 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use setup-ts-deep-modules, ask-matt and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • liufeng1976Skill

    setup-ts-deep-modules

    Wire dependency-cruiser into a TypeScript repo so each package is a deep module — implementation hidden in subfolders, reachable only through its entry-point files. User-invoked.

    Also in 43 projects that use this

  • GoogleSkill

    ask-matt

    Ask which skill or flow fits your situation. A router over the skills in this repo.

    Also in 91 projects that use this

  • RoentekSkill

    loop-me

    Grill me about specs for the workflows I want to build, within this workspace.

    Also in 50 projects that use this

  • AWSSkill

    to-questionnaire

    Turn a decision you can't fully answer into a questionnaire for someone else to fill in.

    Also in 88 projects that use this

  • ryotaroh180105Skill

    claude-handoff

    Hand the current conversation off to a fresh background agent that picks up the work immediately.

    Also in 42 projects that use this

  • GoogleSkill

    to-spec

    Turn the current conversation into a spec and publish it to the project issue tracker — no interview, just synthesis of what you've already discussed.

    Also in 168 projects that use this

  • +3 more from google

The file

We split this skill into the rules it gives an agent

10 rules. 0 are common, 3 are distinctive, and 7 are very specific to this project.

  • 0

    Common

    Most of the 171 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 7

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Follow ../../shared/policy.md.

    line 13
  2. Very specific:

    Run dev-skills:gather-context.

    line 18
  3. Very specific:

    Create it outside the repo: git worktree add ~/.dev-skills/worktrees/<project-slug>/<branch> -b <branch> origin/<base>.

    line 24
  4. Very specific:

    Run dev-skills:spec (which runs dev-skills:qa-plan).

    line 31
  5. Distinctive:

    Add/adjust the automated tests from the QA plan.

    line 38
  6. Very specific:

    Run dev-skills:qa in verify mode against the QA plan and acceptance criteria.

    line 42
  7. Very specific:

    If the change touches an API or documented behavior, run dev-skills:document.

    line 46
  8. Very specific:

    Commit locally per the mode (balanced/guided ask.

    line 49
  9. Distinctive:

    Add to the confirmation list, as applicable:

    line 53
  10. Distinctive:

    Leave the worktree in place.

    line 74

“Common” is measured: kept by at least 6 in 10 of the 171 edited versions of this file we could read. “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

Implement

Follow ../../shared/policy.md. Run dev-skills:project-profile first. Input: a ticket key/URL or a written requirement. With none given and none inferable from the conversation or current branch name, ask for it (the only upfront question).

1. Requirements

Run dev-skills:gather-context. Blockers that make the work impossible: stop and report. Unknowns: handled by the mode inside gather-context.

2. Worktree

  • Branch name: the profile's branch_naming (else <ticket-key-lowercase>-<short-slug>), from the up-to-date base branch (git fetch origin <base>).
  • Create it outside the repo: git worktree add ~/.dev-skills/worktrees/<project-slug>/<branch> -b <branch> origin/<base>. Reuse it if it already exists for this ticket.
  • Copy the main checkout's env files (profile env_files) into it.
  • From here, every command runs in the worktree (absolute paths or cd <worktree> && ...). The user's main checkout is never touched.

3. Plan

Run dev-skills:spec (which runs dev-skills:qa-plan). For a small, obvious change (one or two files, no contract or data change), a short plan in spec.md is enough; still write the test cases. In guided, both are approved before coding.

4. Code

Smallest change that solves the ticket at the root cause, in the surrounding code's style, reusing existing helpers. Whole ticket scope; if part is blocked, do the rest and report what is left. Add/adjust the automated tests from the QA plan. Comments only for non-obvious code. Out-of-plan edits, new dependencies, migrations, deletions: per the mode.

5. Verify

Run dev-skills:qa in verify mode against the QA plan and acceptance criteria. Fix what the change broke and re-run. Never claim green on unverified work.

6. Document

If the change touches an API or documented behavior, run dev-skills:document.

7. Commit

Commit locally per the mode (balanced/guided ask; autonomous commits). Message: the repo's convention from git log, referencing the ticket key.

8. Queue what leaves the machine

Add to the confirmation list, as applicable:

  • Push the branch.
  • Staging deploy, when the profile says this project deploys feature branches to staging (dev-skills:deploy staging, or the project skill the profile names).
  • Open the PR: target = base branch; title and body following the profile's PR template (else: what changed and why, how it was tested, risks, ticket link). Show the full body.
  • Ticket transition/comment only if the profile says the team does that on PR open.

Report

<TICKET> — <title>
Worktree: <path>  Branch: <name>  Commit: <sha | not committed>
Changed:
- `file:line` — what and why
Checks: <command> → pass/fail
Criteria: <n/m met> (details from qa)
Left out: <scope skipped and why | none>
Assumed:
- ...
<confirmation list>

Leave the worktree in place. Remove it (git worktree remove) only when the user asks.

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.