Skip to content
Development
Skill

/check

Run every Iron Law stage and the drift detector suite. Use when the user wants the full project health snapshot, a CI gate, or to verify nothing regressed before committing. Add --strict to promote warn-severity drift findings to errors. Activate only when the connected project

From plugin
cladding
1619 skills6 agents1 command1 MCP
Install
$ npx -y skills add qwerfunch/cladding --skill check --agent claude-code

How it fires

How this skill gets triggered: by you, by Claude, or both.

  • Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.Auto-invocation is when the right skill fires by itself at the right moment, driven by a FLOW.md router and a hook, instead of you invoking it by name. It is the difference between a skill being installed and a skill actually getting used.Read the full definition →
  • You can call itInvoke it directly when you want it.
  • Slash command/check

Context preview

The summary Claude sees to decide when to auto-load this skill.

Run every Iron Law stage and the drift detector suite. Use when the user wants the full project health snapshot, a CI gate, or to verify nothing regressed before committing. Add --strict to promote warn-severity drift findings to errors. Activate only when the connected project

SKILL.md

check.SKILL.md
description: Run every Iron Law stage and the drift detector suite. Use when the user wants the full project health snapshot, a CI gate, or to verify nothing regressed before committing. Add --strict to promote warn-severity drift findings to errors. Activate only when the connected project contains spec.yaml or the user explicitly names Cladding; ignore ordinary requests in uninitialized projects.

Cladding check

Run `clad check` from the project root. Runs the 15 Iron Law stages — Type / Lint / Drift / Commit / Arch / Secret / Unit / Coverage / Spec-conformance / Deliverable-smoke / Smoke / Performance / Visual / Audit / UAT — and aggregates the worst exit code.

  • `0` — every stage cleared or skipped clean.
  • `1` — at least one stage actually failed (fix-required).
  • `2` — every result is skip (no fail-required input on the project yet).

`--strict` promotes warn-severity drift findings to error, matching the CI / pre-publish gate. The Drift stage runs every active detector under `src/stages/detectors/` — `npm run build:plugin` Phase D recounts them and writes the integer into each plugin manifest (e.g. `plugins/claude-code/.claude-plugin/plugin.json`), so the number is never hand-maintained.

`--internal` shows stage codes (`stage_1.1`) instead of business names (`Type`). Default is the business-name surface; the audit log keeps internal ids regardless.

clad check
clad check --strict
clad check --internal

Gate economy (tiers)

Pick the cheapest tier that answers your question — the full pre-push suite is expensive and grows with the project:

  • `clad check --tier=pre-commit` — drift / arch / secret only (spec-vs-code, no full unit suite). Use for

fast inner-loop feedback while implementing.

  • `clad check --tier=pre-push --strict` — the full gate (type / lint / unit / cov + drift). This is what

`clad done <featureId>` already runs, so do NOT run it separately right before `clad done` — one authoritative full gate per feature, not two. A GREEN run refreshes `spec/attestation.yaml` with the running Cladding version, strict blocking mode, detector-catalog SHA-256, module hashes, and feature markers. See `docs/feature-cycle.md` § Gate economy.

Read more
Ships withcladding

For an organization to trust AI with its code, three things must hold — trust, traceability, and stability at scale. cladding wraps your AI coding agent: your intent goes in before it writes, and the result is verified against your spec after, so those three are earned, not assumed. First L4 implementation of the Ironclad standard.

Get the whole plugin

Other skills on cladding.