skills/verify-workflow/SKILL.md
Use when you need to validate workflow state schema or error handling.
npx skillsauth add seokan-jeong/team-shinchan team-shinchan:verify-workflowInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
When this skill is invoked, execute immediately. Do not explain.
| Validator | Command | What it checks |
|-----------|---------|---------------|
| workflow-state-schema | cd "${CLAUDE_PLUGIN_ROOT}" && node tests/validate/workflow-state-schema.js | WORKFLOW_STATE.yaml template + stage/transition gates, plus the big-project layer (project-state.schema.json parses, bigproject PROJECT.yaml template present) |
| error-handling | cd "${CLAUDE_PLUGIN_ROOT}" && node tests/validate/error-handling.js | Error handling patterns are properly defined |
| part-numbering | cd "${CLAUDE_PLUGIN_ROOT}" && node tests/validate/part-numbering.js | Multi-part agent files have correct numbering |
| quick-fix-path | cd "${CLAUDE_PLUGIN_ROOT}" && node tests/validate/quick-fix-path.js | Quick-fix (lite mode) workflow path is valid |
docs/workflow-guide.md, hooks/)cd "${CLAUDE_PLUGIN_ROOT}" && node tests/validate/workflow-state-schema.js
Success criteria:
schemas/project-state.schema.json parses and
skills/bigproject/SKILL.md contains a PROJECT.yaml templateOn failure:
cd "${CLAUDE_PLUGIN_ROOT}" && node tests/validate/error-handling.js
Success criteria:
On failure:
cd "${CLAUDE_PLUGIN_ROOT}" && node tests/validate/part-numbering.js
Success criteria:
On failure:
cd "${CLAUDE_PLUGIN_ROOT}" && node tests/validate/quick-fix-path.js
Success criteria:
On failure:
The executing stage is driven by src/dag-executor.js over a task DAG. Each PLAN.md task MUST
carry the six DAG fields (id, depends_on, touches, verify, estimate, scope) and the
plan SHOULD declare integration_test metadata. The executor enforces a per-task verify gate and
a strict ALL-PASS completion gate before the executing → done transition.
cd "${CLAUDE_PLUGIN_ROOT}" && node --test tests/dag-executor.test.js
Success criteria:
src/dag-executor.js exposes parsePlan, topoSort, buildConflictGraph,
connectedComponents, evaluateVerify, recoveryStep, evaluateCompletionGate,
runIntegration, executePlanOn failure:
docs/dag-executor.md)testing
Default-on interview option-quality panel — N diverse generators produce structure-free options, a SelfCheckGPT majority-vote consensus filters hallucinations, a SteerConf cautious-confidence judge scores survivors, and a deterministic top-K is returned. Workflow tier; the single fierce-* skill that is ON by default.
development
Deterministic adversarial code review for high-stakes scope — independent per-dimension review, a non-skippable per-finding refutation, completeness + interaction critics, and a deterministic 3-lens rubric judge panel. Opt-in main-loop Workflow tier.
data-ai
Deterministic loop-until-done for high-stakes long-running tasks — a worker/verifier loop the script bounds by iteration cap, token budget, and stagnation, closed by an Action-Kamen gate. Opt-in main-loop Workflow tier.
testing
Deterministic adversarial debate for high-stakes or irreversible decisions — mandatory refutation plus a scored judge panel. Opt-in main-loop Workflow tier.