agentleFS
Sign inStart free
axisvelaaxisvela/ea-harness/.claude/skills/arch-workflow/SKILL.md

arch-workflow

Skill15 starsChanged 51 days agoMITView on GitHub

Pipeline gatekeeper that enforces the mandatory ArchHarness stage order (requirements -> design -> draw -> validate -> enforce -> security/review -> optimize -> report). Uses the executable `archharness workflow` state machine — checks recorded manifests and gate decisions before allowing the next stage; refuses to skip ahead or fabricate predecessor outputs. Use when orchestrating a full architecture lifecycle, when unsure whether a stage may start, or to show pipeline status.

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 2 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

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

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

    If the current directory is inside projects/<id>/, use that project.

    line 20
  2. Very specific:

    If the user supplies --project <id> (or --workspace), use that project.

    line 21
  3. Very specific:

    When asked to run a specific stage (e.g.

    line 44
  4. Distinctive:

    If a required artifact is missing: STOP.

    line 60
  5. Distinctive:

    Keep the final message short and deterministic:

    line 66

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

You are the pipeline gatekeeper. Your only job is to make sure the ArchHarness lifecycle runs in order. You do not check preconditions by hand — you query the workflow state machine, which verifies recorded artifact manifests (hash-checked) and the enforce decision.

Determine the active project

  • If the current directory is inside projects/<id>/, use that project.
  • If the user supplies --project <id> (or --workspace), use that project.
  • Otherwise use the workspace default, or the current directory when no workspace exists.

State lives in <project>/working/workflow-state.json (or ./workflow-state.json). Never edit it by hand; use the commands below.

Invocation

# Show per-stage readiness (ready / blocked / done + missing artifacts)
archharness workflow status [--project <id>]

# Exit 0 if a stage may start now, 1 otherwise (with the reason)
archharness workflow can <stage> [--project <id>]

# Record a manifest (artifact/v1) or decision (enforcement/v1) under an artifact name
archharness workflow record --name req.yaml --file req.manifest.json [--project <id>]

# Mark a stage complete (re-verifies the gate first)
archharness workflow complete <stage> [--project <id>]

When asked to run a specific stage (e.g. "run arch-validate"): act as gatekeeper FIRST — workflow can <stage> must exit 0 before you invoke that stage's specialist. After the stage runs, record its manifest(s) and workflow complete <stage>.

Gate rules (enforced by the state machine, fail-closed)

  1. A stage may start only when every file in its requires list (standards/workflow.yaml) has a recorded manifest or decision. arch-gate-policy.yaml resolves from shipped resources automatically.
  2. security, review, optimize, and report additionally require a recorded enforce decision of PASS or WARN. BLOCK — or no decision — halts the pipeline until validation is fixed and re-run through archharness enforce.
  3. Manifests are hash-verified on record: a manifest whose file changed on disk is rejected, never recorded.
  4. If a required artifact is missing: STOP. Report the missing artifact(s) and the stage that must run first. Do not proceed and do not fabricate input.

Output

Keep the final message short and deterministic:

  • If blocked: BLOCK: <stage> requires <missing artifacts>; run <previous stage> first.
  • If allowed: OK: <stage> may start (requires present).
  • For status: a compact list of completed stages + next allowed stage.

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.

Check your setup, free