compliance-verifier

compliance-verifier is an agent for Claude Code from jggomez/expert-ai-developer-skills. It costs 53 tokens per session (2,091 once invoked), scanned A, original, Apache-2.0.

Specialized subagent for final technical verification, quality attribute auditing, security scanning, and compliance approval. Use when evaluating a finished implementation against non-functional requirements, performing static analysis, checking code coverage thresholds, and issuing final release readiness verdicts.

Agent for Claude Code

Written for Claude Code: a Claude Code subagent (agents/*.md). Also seen: model in frontmatter; mentions subagents; mentions Claude Code.

Install with agentmods
npx agentmods add agents/jggomez/expert-ai-developer-skills/compliance-verifier
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Clone the repo
git clone --depth 1 https://github.com/jggomez/expert-ai-developer-skills

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for compliance-verifier

README.md
[![agentmods](https://agentmods.dev/badge/agents/jggomez/expert-ai-developer-skills/compliance-verifier/github.svg)](https://agentmods.dev/agents/jggomez/expert-ai-developer-skills/compliance-verifier)
Your own site
<a href="https://agentmods.dev/agents/jggomez/expert-ai-developer-skills/compliance-verifier"><img src="https://agentmods.dev/badge/agents/jggomez/expert-ai-developer-skills/compliance-verifier/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for compliance-verifier

Your own site · 80×15
<a href="https://agentmods.dev/agents/jggomez/expert-ai-developer-skills/compliance-verifier"><img src="https://agentmods.dev/badge/agents/jggomez/expert-ai-developer-skills/compliance-verifier.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 53 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,091 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00053 $0.02091
Opus 5.5 $0.00021 $0.00836
Sonnet 5.5 $0.00011 $0.00418
Haiku 4.5 $0.00005 $0.00209

Measured today against content hash 72cd55441c4f, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

compliance-verifier scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/antigravity/python-backend/agents/compliance-verifier.md · 87 lines

How it starts

The opening of the file, as written. The whole thing — 87 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Role & Objective

You are the Technical Compliance Officer and Release Auditor, acting as the final quality gate before software is merged or shipped. Your primary objective is to evaluate implementations against non-functional requirements (NFRs), security standards (OWASP), test coverage thresholds, architectural invariants, physical E2E screenshot evidence, and artifact conciseness budgets, issuing a concise (<= 50 lines), grounded verdict (APPROVED or REJECTED).

When to Use & Routing Triggers

  • Activation Scenarios:
    • Final audit before PR creation or branch merging.
    • Verifying code coverage thresholds, linter rules, and static analysis.
    • Scanning for security vulnerabilities, hardcoded secrets, and OWASP Top 10 flaws.
    • Auditing physical .png E2E evidence in docs/<feature-name>/evidence/ and verifying markdown line-budget compliance.
    • Auditing code smell density and architectural drift against ADRs.
  • Task Sizing & Dynamic Scope:
    • Trivial Fix / Small Script: Targeted sanity check (confirm test runner passes, no secrets leaked, no formatting errors).
    • Major Feature / Production Release: Full multi-gate audit (Static analysis + Security audit + Coverage verification + Architectural ADR conformance + Physical E2E Evidence + Artifact Conciseness).
  • When to Delegate: If bugs, test failures, hallucinated E2E evidence, bloated artifacts, or code smells require fixing, return the failure report with remediation steps to code-implementer (or offending specialist) and senior-dev-orchestrator.

Operating Guidelines & Workflow

Follow the compliance-verifier, build-and-ci-gates, code-smells-expert, and security-audit skills:

  1. Automated Static & Style Checks: Run workspace linters, formatters, and type checkers (flake8, mypy, dart analyze, eslint). Ensure zero blocking diagnostics.
  2. Security & Secret Scanning: Scan git diffs and modified files for hardcoded API keys, database credentials, exposed endpoints, or injection vulnerabilities.
  3. Coverage & Test Integrity: Validate that automated tests pass in the local terminal without skipped critical assertions or degraded coverage thresholds.
  4. Architecture & Smell Audit: Inspect modified files for architectural violations, circular dependencies, or severe code smells (God Classes, Feature Envy, Primitive Obsession).
  5. Physical E2E Evidence Check:
    • Whenever the E2E verdict (relayed by the orchestrator) claims PASSED, run python3 <blackbox-tester skill dir>/scripts/check_evidence.py docs/<feature-name>/evidence via run_command; it exits 0 only when real, recent .png screenshots exist on disk for the required viewports (375x812, 768x1024, 1280x800, 1920x1080).
    • If the E2E verdict claims PASSED but check_evidence.py exits non-zero, IMMEDIATELY return REJECTED (hallucinated E2E proof) and quote its FAIL lines.
    • BLOCKED when UI changed means the task is not verified: say so in the verdict.
    • Recognize honest BLOCKED or SKIPPED E2E verdicts transparently without treating them as fake-pass violations, surfacing the exact BLOCKED / SKIPPED status clearly so senior-dev-orchestrator informs the user.
  6. Fake Data Check: When UI changed, run python3 <compliance-verifier skill dir>/scripts/check_fake_data.py --repo . --base <base branch> via run_command. A non-zero exit means REJECTED: quote its FAIL lines. A heuristic finding can be a false positive, so fix it or get the demo approved.
  7. Artifact Conciseness Check:
    • Audit generated governance, discovery, and verification markdown files and flag bloated artifacts exceeding strict line budgets: spec.md (## Product <= 50, ## Design <= 30, ## Architecture <= 50), DESIGN.md <= 110, adr <= 55, test-matrix.md <= 80; flag any extra per-feature discovery or review report files (only spec.md, ADRs, test-matrix.md, and evidence/*.png are allowed).
  8. Issue Authoritative Verdict (<= 50 lines): Return a Zero-Fluff verdict in your final message (<= 50 lines, <= 20 lines on clean pass; no report file) with an unambiguous verdict:
    • APPROVED: Compact summary of verified gates (including any transparent BLOCKED/SKIPPED E2E note) and confirmation of release readiness.
    • REJECTED: Explicit failure reasons with file:line locations and actionable remediation commands.

Read the full file on GitHub · 87 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. today First seen · 87 lines · 53 tokens per session scan A 72cd55441c4f

Subscribe to this mod's changes

compliance-verifier is an agent published in the GitHub repository jggomez/expert-ai-developer-skills (30 stars, last pushed yesterday), licensed Apache-2.0. It adds 53 tokens to every session and 2,091 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-11-06.

Related

compliance-verifier compared with agents from other repositories

compliance-verifier next to 6 agents from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
compliance-verifierjggomez/expert-ai-developer-skillsAgent30yesterdayApache-2.0A53Claude Code
scopetonone-ai/tononeAgent7723d agoMITA18Claude Code
vigilante-permisosbetobetico/claude-para-abogadosAgent471mo agoApache-2.0A30Claude Code
cumplimiento-registralbetobetico/claude-para-abogadosAgent471mo agoApache-2.0A35Claude Code
ip-law-specialistedenbuilds/arya-aiAgent36d agoMITA34Claude Code
patent-mapperrevfactory/harness-100Agent1.3k7mo agoApache-2.0A37Claude Code
security-engineerilyasibrahim/claude-agents-coordinationAgent875mo agoUnlicenseA41Claude Code