adlc-verify - SKILL.md Agent Skill

name: adlc-verify description: Verify implementation completeness against ADLC plans, rules, and repo behavior.

ADLC Verify

Use this after implementation and before review or commit.

Process

Resolve effective ADLC paths from .adlc/config.yaml; use adlc resolve-config when the CLI is available.
Read the active plan, any cited interview spec, configured rules, architecture, and git diff.
Map each plan task and acceptance criterion to evidence, including interview acceptance criteria when present.
Run the strongest practical repo checks.
Look for leftover TODOs, env/config drift, skipped tests, stale docs, and plan-vs-code gaps.
Classify failures as autonomous-fixable or human-gated. Code, docs, tests, rules, and security gaps should suggest adlc-fix; human-gated blockers are limited to explicit decisions, credentials or external accounts, destructive or production operations, legal/security sign-off, scope or product ambiguity, or user-requested approval points.
Use strict mode before merge or release.

Output

End with a final parseable adlc-gate-result fenced block:

{
  "schema_version": 1,
  "gate": "verify",
  "status": "pass|warn|fail",
  "blocking": false,
  "blockers": [],
  "affected_files": [],
  "suggested_next": {
    "command": "adlc-review|adlc-fix|null",
    "reason": "Short reason."
  }
}

Use suggested_next.command: "adlc-fix" for autonomous-fixable blockers. Use null only when a human-gated blocker prevents the next autonomous gate.