auditor-gate
Apply final governance and release-gate checks to a judged change set by reading `handoff.json`, `verdict.json`, optional eval evidence, and emitting machine-readable `audit.json` with `gate` status. Use when implementation already has a judge verdict and a separate auditor must decide landability (`pass`, `fail`, or `needs-human`) without modifying source files.
Best use case
auditor-gate is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Apply final governance and release-gate checks to a judged change set by reading `handoff.json`, `verdict.json`, optional eval evidence, and emitting machine-readable `audit.json` with `gate` status. Use when implementation already has a judge verdict and a separate auditor must decide landability (`pass`, `fail`, or `needs-human`) without modifying source files.
Teams using auditor-gate should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/auditor-gate/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How auditor-gate Compares
| Feature / Agent | auditor-gate | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Apply final governance and release-gate checks to a judged change set by reading `handoff.json`, `verdict.json`, optional eval evidence, and emitting machine-readable `audit.json` with `gate` status. Use when implementation already has a judge verdict and a separate auditor must decide landability (`pass`, `fail`, or `needs-human`) without modifying source files.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# Auditor Gate Run final acceptance gating as a separate role after judge evaluation. ## Role Boundaries - Audit process and risk posture. - Validate traceability and policy signals. - Produce `audit.json` only. - Do not implement fixes and do not re-grade functional correctness from scratch. ## Required Inputs - `task_id` - `handoff.json` - `verdict.json` - output path for `audit.json` Optional: - `eval-results.json` - explicit audit commands (policy/security/license/quality checks) ## Workflow ### 1) Run audit evidence collection Run: ```bash python3 <path-to-skill>/scripts/run_audit.py \ --repo <repo-root> \ --handoff <artifact-path>/handoff.json \ --verdict <artifact-path>/verdict.json \ --output <artifact-path>/audit-results.json \ --command "python3 -m pip list --format=freeze" ``` This collects artifact-integrity checks and command outcomes. It also checks task and requirement traceability between `handoff.json` and `verdict.json`. ### 2) Write gate artifact Run: ```bash python3 <path-to-skill>/scripts/write_audit.py \ --task-id <task-id> \ --audit-results <artifact-path>/audit-results.json \ --output <artifact-path>/audit.json \ --gate fail \ --finding "policy::Dependency policy review missing" \ --required-action "Add dependency review evidence" \ --risk-level high ``` ### 3) Validate gate artifact Run: ```bash python3 <path-to-skill>/scripts/validate_audit.py \ --input <artifact-path>/audit.json ``` ## Decision Rules - `pass`: - no critical audit failures - traceability and policy checks pass or are explicitly accepted - `fail`: - policy, traceability, or release constraints are violated - `needs-human`: - missing evidence or ambiguous policy interpretation blocks deterministic decision ## Output Rules - Always produce `audit.json`. - Keep findings evidence-based and implementation-neutral. - Provide concrete `required_actions` for `fail` and `needs-human`. - Report requirement-coverage traceability explicitly in `audit.json`. ## Resources ### scripts/ - `scripts/run_audit.py`: collect artifact and command-level audit evidence. - `scripts/write_audit.py`: synthesize `audit.json` gate decision. - `scripts/validate_audit.py`: validate audit artifact schema. ### references/ - `references/artifact-contract.md`: canonical `audit.json` structure. - `references/gate-rubric.md`: practical criteria for `pass`/`fail`/`needs-human`.
Related Skills
auditor-workflow
Group-level implementation audit workflow for auditor agents. Handles loading project rules, reading connected phases, reviewing code reviews, checking deferred items, cross-phase impact analysis, verification, and structured reporting to the orchestrator. Invoke this skill as your first action — not user-invocable.
architecture-auditor
Architecture audit and analysis specialist for Modular Monoliths. **ALWAYS use when reviewing codebase architecture, evaluating bounded contexts, assessing shared kernel size, detecting "Core Obesity Syndrome", or comparing implementation against ADR-0001 and anti-patterns guide.** Use proactively when user asks about context isolation, cross-context coupling, or shared kernel growth. Examples - "audit contexts structure", "check shared kernel size", "find cross-context imports", "detect base classes", "review bounded context isolation", "check for Core Obesity".
AI Safety Auditor
Audit AI systems for safety, bias, and responsible deployment
ai-doc-system-auditor
No description provided.
agent-security-auditor
Expert security auditor specializing in comprehensive security assessments, compliance validation, and risk management. Masters security frameworks, audit methodologies, and compliance standards with focus on identifying vulnerabilities and ensuring regulatory adherence.
agent-compliance-auditor
Validates agent definitions against the Antigravity audit rubric.
Accessibility Auditor
Web accessibility specialist for WCAG compliance, ARIA implementation, and inclusive design. Use when auditing websites for accessibility issues, implementing WCAG 2.1 AA/AAA standards, testing with screen readers, or ensuring ADA compliance. Expert in semantic HTML, keyboard navigation, and assistive technology compatibility.
approval-gate
ワークフローの重要なフェーズ移行前にユーザーの明示的な承認を必要とする承認ゲートの共通フォーマットとパターンを定義
ai-search-technical-auditor
Audit front-end code for AI search readiness. Use when reviewing HTML structure, meta tags, schema markup, and technical elements that affect how AI crawlers understand and index web pages.
quality-gate
Synthesize all quality assessments (risk, test-design, traceability, NFR) into evidence-based gate decision (PASS/CONCERNS/FAIL/WAIVED) with comprehensive rationale. Generates both YAML (CI/CD) and Markdown (human review) reports with action items. Use during final quality review to make go/no-go deployment decisions based on comprehensive quality evidence.
50-execute-gated-150
[50] EXECUTE. Execute plans step-by-step with confirmation gates. Each step requires user approval before proceeding. Includes change management lifecycle (Pre-Change → During → Post-Change → Rollback). Use when implementing approved plans, deploying changes, or any multi-step execution requiring control and reversibility.
vibe-code-auditor
Audit rapidly generated or AI-produced code for structural flaws, fragility, and production risks.