agentleFS
KcAnomKcAnom/pi-fairy-tales-chains/skills/feature-ship-review/SKILL.md

feature-ship-review

Skill0 starsChanged 4 months agoMITView on GitHub

Phase 3 of 4 in the feature-ship chain (spec → build → review → ship). Runs the deep-review skill and fairy-tales review-role subagents against the Phase 2 diff, collecting blocking and non-blocking findings for Phase 4 to act on. Trigger when the previous phase says run feature-ship-review next, or standalone via "feature-ship-review", "phase 3 of feature-ship", "review the build", "review this feature".

The file

We split this skill into the rules it gives an agent

11 rules. About 0 are common, 3 are distinctive, and 8 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 8

    Very specific

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

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

    All chain state is managed by the chain tool: authoritative JSON at .pi/fairy-tales/chains/feature-ship/state.json with a human-readable …

    line 26
  2. Very specific:

    Call the chain tool with action: "status", chain: "feature-ship" (a legacy .feature-ship-state.md from an older version is imported …

    line 33
  3. Very specific:

    If there is no run → abort: "No feature-ship run found.

    line 36
  4. Very specific:

    If the run is not active with currentPhase: "review" → abort: "Expected the run to be at phase review but it is at <currentPhase/status>.

    line 37
  5. Very specific:

    If the build output (data.filesChanged / the build phase's artifacts.branch or summary) is missing → abort: "Phase pointer looks right but …

    line 40
  6. Very specific:

    If the tool reports the chain is locked by another session, abort.

    line 43
  7. Very specific:

    If a returned quest is already done (crash-resume case), reuse its stored result.

    line 73
  8. Very specific:

    After extracting findings into chain state (Step 5), quest action consume each quest by id.

    line 75
  9. Distinctive:

    Call the chain tool:

    line 98
  10. Distinctive:

    After completing this phase, output exactly:

    line 115
  11. Distinctive:

    Do not start the next phase.

    line 120

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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

Feature Ship: Review

Chain version: 2 (durable state contract — see docs/STATE-CONTRACT.md)

Phase 3 of 4: feature-ship-spec → feature-ship-build → feature-ship-review → feature-ship-ship

Overview

Reviews the diff produced in Phase 2 against the spec's acceptance criteria from Phase

  1. Runs the deep-review skill for a structured multi-angle pass over the diff, and dispatches fairy-tales review-role subagents for targeted checks (correctness, security, spec conformance), then merges everything into one findings list split into blocking and non-blocking so Phase 4 knows exactly what must be fixed before shipping.

All chain state is managed by the chain tool: authoritative JSON at .pi/fairy-tales/chains/feature-ship/state.json with a human-readable state.md projection beside it. Never create or edit state files by hand, and never create lock files — the tool locks for you.

Load State

Call the chain tool with action: "status", chain: "feature-ship" (a legacy .feature-ship-state.md from an older version is imported automatically):

  • If there is no run → abort: "No feature-ship run found. Run feature-ship-spec first."
  • If the run is not active with currentPhase: "review" → abort: "Expected the run to be at phase review but it is at <currentPhase/status>. Run that phase's skill instead."
  • If the build output (data.filesChanged / the build phase's artifacts.branch or summary) is missing → abort: "Phase pointer looks right but the Phase 2 output is missing. Treat as corrupted — restart the chain or repair via chain action 'update'."
  • If the tool reports the chain is locked by another session, abort; a dead session's lock can be cleared with action: "unlock". Never create lock files yourself.

Note the run's runId — it keys this phase's quest dedupe keys.

Workflow

Step 1 — Assemble the diff

Get the exact diff to review from the location Phase 2 recorded (branch, worktree, or patch file — artifacts.branch / the build phase summary). Do not re-derive it from the spec — review what was actually built.

Step 2 — Run the deep-review skill

Invoke the deep-review skill against the diff. Let it run its full multi-angle pass (correctness, edge cases, consistency) rather than summarizing it yourself.

Step 3 — Dispatch review-role subagents via durable quests

Dispatch fairy-tales review-role subagents in parallel, each with a narrow brief. Because review passes can be long-running and must survive a session restart, use the quest tool rather than a bare agent call — one quest per subagent:

  • quest action enqueue with role: "review", the brief below, and:
    • dedupeKey: "feature-ship/<runId>/review/<unit>" (e.g. acceptance, security, or a risk area name from the Phase 1 spec) — re-running a crashed phase returns the same quest instead of duplicating a review pass;
    • chain: { chain: "feature-ship", runId: "<runId>", phase: "review" };
    • retainUntilConsumed: true.
  • If a returned quest is already done (crash-resume case), reuse its stored result; otherwise quest action run with its id and collect the result.
  • After extracting findings into chain state (Step 5), quest action consume each quest by id.

Briefs to fan out:

  • One checks the diff against the Phase 1 acceptance criteria specifically — does every criterion have code behind it?
  • One checks for correctness/security issues the deep-review skill's pass might not specialize in (e.g., auth, injection, data handling) if the diff touches that surface.
  • Additional quests as the diff's risk areas (from the Phase 1 spec's "Risks" section) warrant.

Step 4 — Merge findings

Combine the deep-review skill's output and every review-role quest's output into one list. Classify each finding as:

  • Blocking — violates an acceptance criterion, introduces a correctness/security bug, or breaks something that worked before.
  • Non-blocking — style, minor cleanup, "nice to have" — safe to ship without.

Step 5 — Complete the phase

Call the chain tool:

action: "complete-phase", chain: "feature-ship", phase: "review",
summary: "deep-review skill + review-role quest findings merged: <N> blocking, <M> non-blocking",
data: {
  "blockingFindings": ["<finding 1>", "<finding 2>", "..."],
  "nonBlockingFindings": ["<finding 1>", "..."]
},
artifacts: { "reviewQuests": "<quest ids>" }

The tool validates this is the current phase, advances the run to ship, and rewrites the JSON + markdown projection atomically.

Handoff

After completing this phase, output exactly:

Phase 3 complete. Chain state updated (.pi/fairy-tales/chains/feature-ship/state.json). Run /feature-ship-ship to begin Phase 4.

Do not start the next phase. Do not offer to continue. Output only the handoff line and stop.

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.