agentleFS
lifemixturelifemixture/ai-orchestration-repo/.claude/skills/feature-build-step-by-step/SKILL.md

feature-build-step-by-step

Skill0 starsChanged 3 months agoMITView on GitHub

Implement a planned multi-repo feature with a stop before and after every phase, so the user reviews each diff as it lands instead of all at once. Same engine as /feature-build — implementer and independent reviewer as sibling subagents, tests as the evidence a phase is done — with every gate switched on. Commits; never pushes.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 1 is distinctive, and 1 is 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.

  • 1

    Very specific

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

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

    Gate policy: every-phase: All four gates are on: brief before each phase (G1), confirm after each phase (G2), stop on any surprise (G3), …

    line 12
  2. Distinctive:

    Use this when the user wants to review the work incrementally rather than face one large diff at the end.

    line 15

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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

You are implementing a planned feature across the repositories it touches, with the user in the loop at every phase. Feature folder: $ARGUMENTS

Gate policy: every-phase. All four gates are on: brief before each phase (G1), confirm after each phase (G2), stop on any surprise (G3), report at the end (G4).

Use this when the user wants to review the work incrementally rather than face one large diff at the end. Two things matter more here than in the other variant:

  • At G2 the user reads the diff, not your description of it. Give them the command and wait. A summary written by the side that wrote the code is not a review, and you have not seen the code either — that is by design.
  • Offer the ladder. At any gate, the user may say "continue without asking" for the rest of this service or the rest of the feature. Honour it by switching to the on-surprise policy from that point. Trust is earned during a run, and they should not have to restart under a different skill to act on it.

Now read and follow ${CLAUDE_PROJECT_DIR}/.claude/lib/build-engine.md, applying the every-phase policy at its gate points. Everything else — roles, the review loop, the context rule, commits, the checklist — is defined there and is identical across the variants.

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.