agentleFS
christofferbergjchristofferbergj/agent-skills/skills/engineering/implement-spec/SKILL.md

implement-spec

Skill1 starsChanged 21 days agoMITView on GitHub

Implement the result of /to-spec and /to-tickets in code.

107 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use implement-spec-with-codex, close-spec 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.
  • dzhngSkill

    implement-spec-with-codex

    Run [implement-spec](../implement-spec/SKILL.md) with Codex doing the implementation passes while you orchestrate, integrate, and review. Use when the user asks to implement a spec with Codex, or to run implement-spec with Codex as the implementer.

    Also in 4 projects that use this

  • dzhngSkill

    close-spec

    When a spec under specs/<feature>/ is done shipping (a write-spec build, or any planned task that produced a spec), archive it to specs/done/ and rewrite it from a build-plan into a durable rationale — the why, the principles, the invariants — pointing back to the real code for the how. Use when implementation has landed and the plan no longer matches what shipped, or the user says a feature/spec is finished. Pairs with [write-spec](../write-spec/SKILL.md) (the plan this closes) and [review](../review/SKILL.md) (run before closing).

    Also in 4 projects that use this

  • wolverin0Skill

    audit-decisions

    Standardize remediation decisions, baseline-policy entries, and legal-review banners for accepted, deferred, or policy-blocked audit findings.

    Also in 4 projects that use this

  • saffron-healthSkill

    get-pr-reviews

    Get nicely formatted output of the last review comments on a PR. Used internally by other skills.

    Also in 3 projects that use this

  • saffron-healthSkill

    push-everything

    Stage ALL changes in the repository (not just session changes), commit, and push. Use when you need to commit all pending changes.

    Also in 3 projects that use this

  • transitive-bullshitSkill

    compare-screenshots

    Compare screenshots against the intended design, distinguishing approved references from historical baselines. Use for iterative implementation matching, before/after visual review, objective image telemetry, or checking a lone capture for flat, empty, or badly framed content.

    Also in 4 projects that use this

  • +5 more from dzhng

The file

Files
SKILL.md

Implement a spec

Implement the entire spec on one integration branch, treating its tickets as a dependency graph. One coordinator owns merges, canonical project artifacts, and tracker updates.

  1. Read the full spec, tickets, blocking edges, existing branch state, and configured tracker. Reuse the current source of truth. If no tracker is configured, tell the user to run /setup-matt-pocock-skills; until then, local artifacts can carry the graph without external publication. Invoke /domain-modeling to resolve project docs and /principles for implementation before code changes. Complete when scope, accepted seams, branch base, ready frontier, and any declared migration verification units are explicit and the graph has no unresolved cycle.
  2. Create or reuse an owned integration branch. When the tracker closes work through PRs or the user asks for one, and publication is authorized, open a draft PR marked as closing the spec and tickets after the first merge in step 3. Gather bounded shared exploration once when useful, saving its notes outside the repository where every writer can read them. Each ready ticket runs through /pstack-mode's playbook that fits on two terms: the ticket's accepted design is /architect's chosen shape, so no second design round runs, and the playbook stops at a verified commit on the ticket's branch, since step 4's single /review-loop and step 5's PR replace its Opening a PR. Invoke /swarm to dispatch ready tickets within host capacity, each to a fresh writer on its own worktree based on the verified integration revision. In T3 Code the writer is a T3 child task on /swarm's T3 route whose brief says to follow the installed pstack-mode SKILL.md on those terms, since a native subagent can't run a playbook that spawns subagents of its own; elsewhere, run each ticket's playbook from this chat and delegate its writer step through /swarm. Point writers to the spec, tickets, notes, and commits instead of duplicating them, and explicitly delegate /tdd, /principles for implementation, resolved principle instructions, accepted design decisions, and acceptance criteria. Each writer merges the integration tip into its branch before reporting done. Inspect a wrong worktree base and preserve its changes before repair; never reset unknown work. Complete when every running ticket has an owner, base, playbook, applicable principle owners, and evidence contract.
  3. Inspect each terminal result, merge accepted changes serially, and update the frontier only after the declared ticket or migration-phase checks pass. For a coupled migration, use /principle-migrate-callers-then-delete-legacy-apis for the unit's completion and dependent-work gates. Writers return proposed domain edits; reconcile them through /domain-modeling before canonical docs change. Use the existing tracker graph and link execution evidence instead of duplicating tickets. Complete each ticket only at its declared verification boundary with its integrated acceptance evidence.
  4. When all required tickets are integrated, recheck /principles for verification, run relevant whole-spec checks, and invoke /review-loop once at the final revision, passing it any review evidence that covers this exact artifact. After its fixes, recheck affected whole-spec behavior. Complete when the entire spec is demonstrated against the actual artifact or the validated subset and remaining blockers are explicit.
  5. Deliver a draft PR by default when authorized, using /pr and reusing one opened in step 2. Keep readiness, ticket closure, and merging at their explicitly authorized project boundaries; code on an unmerged branch is not automatically shipped. Use native worktree archive/removal for owned managed checkouts after inspecting preservation needs. Complete when all workers are accounted for, evidence survives, and no unknown work is removed.

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.