More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Legal, Security & Compliance Auditor

    2

    Adaptive GDPR, CCPA, security, and AI compliance audit with severity-graded findings and law citations

    $10

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned

    Legal, Security & Compliance Auditor

    Example session with this skill installed

    I'm launching a SaaS app (Next.js + Prisma + Clerk + OpenAI) in the EU and California next month. Run a full privacy
    and compliance audit and tell me what to fix first.

    • Read your context and instructions
    • Compiled the legal security

    Scope determined: GDPR + UK GDPR + CCPA/CPRA + Security + LLM. LGPD/PIPL/DPDPA not in scope.

    I scanned 142 source files. Writing the full report to COMPLIANCE_AUDIT.md.

    Summary: 20 findings — 3 Critical, 6 High, 7 Medium, 4 Low.

    Top priority right now: an OpenAI API key was committed in PR #214 and never rotated. The key is reachable via
    git log --all -S 'sk-proj-' even though it was removed in a later commit. Rotate it in the OpenAI dashboard now,
    then install gitleaks as a pre-commit hook.

    Other Critical findings:

    • Request middleware logs full POST bodies — passwords leak to Sentry on signup (src/middleware.ts:18).
    • No account-deletion endpoint exists (GDPR Art. 17, CCPA §1798.105). The schema has no anonymization helpers either.

    Recommended order: fix the 3 Criticals this week. Add the EU cookie consent gate and "Do Not Sell or Share" link
    before launch. The full report at COMPLIANCE_AUDIT.md contains file:line for each finding, the article cited, and a
    concrete fix.

    This audit is an engineering checklist, not legal counsel. Consult a qualified data protection lawyer for binding
    decisions.

    142 source files. Writing the full report to COMPLIANCE_AUDIT.md

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Ship to production with confidence. This skill runs a deep privacy and security audit of your codebase and produces a single, prioritized report — like having a junior data-protection engineer on call.

    What it audits

    • GDPR + UK GDPR — 23 checks: lawful basis, consent, all data-subject rights (Art. 15–22), data protection by design, security, breach notification, international transfers
    • CCPA / CPRA — 17 checks: notice at collection, sale/share opt-out, Global Privacy Control, minor opt-in, sensitive PI, request response SLA
    • Data security — 20 checks: password hashing, encryption at rest and in transit, secrets in code, PII in logs, JWT and session security, SQL injection, security headers, CORS
    • LLM disclosure & AI compliance — 16 checks: AI labeling, inaccuracy warnings, data sent to providers, EU AI Act Art. 50, prompt injection, vector store privacy, automated decisions
    • LGPD (Brazil), PIPL (China), DPDPA (India) — 24 conditional checks loaded only when the skill detects you target those markets

    How it works

    1. Detects target markets from i18n files, currencies, domains, languages, framework signals. Asks if signals are ambiguous.
    2. Maps your PII surface — database models, API endpoints, forms, logs, third parties, LLM integrations, cookies, auth flows.
    3. Runs applicable checklists against the map.
    4. Writes COMPLIANCE_AUDIT.md with Critical / High / Medium / Low findings. Every finding has the exact file:line, the article cited (e.g. "GDPR Art. 17(1)(a)"), what's wrong, why it matters, and a concrete fix.

    Why this skill stands out

    • Adaptive scope — auto-detects target jurisdictions, so you get only the checks that apply
    • Concrete fixes, not vague advice — "Replace SHA-256 with argon2id (m=19456 KB, t=2, p=1), migrate on next login" beats "use stronger hashing"
    • Honest boundaries — produces an engineering audit, not legal advice; report ends with disclaimer recommending qualified counsel
    • Agent-agnostic — Claude Code, Cursor, Codex CLI, VS Code Copilot, Gemini CLI, Windsurf
    • Stack-aware — recognizes Prisma, Drizzle, TypeORM, SQLAlchemy, Django ORM, Active Record, Mongoose, plus Next.js, FastAPI, Express, NestJS, Rails routing

    Who it's for

    • Founders preparing for production launch in EU / UK / US
    • Engineers preparing for a SOC 2 / ISO 27001 / DPIA review
    • Teams adding AI features who need to disclose properly
    • Anyone shipping a product that touches user data

    Trigger phrases

    The skill auto-loads when you naturally ask things like:
    • "Check my app for GDPR compliance"
    • "Do a privacy audit before launch"
    • "Find data protection issues"
    • "Audit my codebase for personal data handling"
    • "Review my app before launching in the EU"

    Limitations

    Static review only. Does not write fixes (you decide what to change), does not generate Privacy Policy / Terms text, does not certify compliance. Consult a qualified data protection lawyer for binding decisions.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Security.

    Static analysis only — does not detect issues that only surface at runtime (e.g., race conditions in consent flows, certain side-channel leaks).
    Does not write code fixes — produces findings and concrete recommendations; you decide what to change.
    Does not generate Privacy Policy or Terms of Service text. Use a vetted template or a qualified lawyer for those.
    Does not certify compliance. Only a qualified DPO and lawyer can do that.
    Pattern recognition is tuned for modern web and mobile stacks. Less common languages (Elixir, Clojure, OCaml, COBOL) fall back to generic semantic checks with reduced precision.
    Law content reflects 2025 baselines (CPRA enforcement, EU AI Act in force, DPDPA Draft Rules of January 2025, CAC 2024 transfer clarifications). Major regulatory shifts after mid-2025 may require a skill update.
    Conditional jurisdictions (LGPD/PIPL/DPDPA) trigger only on detected market signals; if signals are weak the skill asks before applying them.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 24 days ago

    Software
    Git

    Permissions

    Read Files
    Write Files

    File Scopes

    **/*.{ts
    tsx
    js
    jsx
    mjs
    cjs}
    **/*.{py
    rb
    go
    java
    cs
    php
    rs
    swift
    kt}
    **/*.{json
    yaml
    yml
    toml
    xml}
    **/.env*
    **/schema.prisma
    **/migrations/**
    **/legal/**
    **/privacy*
    **/terms*
    COMPLIANCE_AUDIT.md

    This skill performs static analysis only. It reads source files to detect PII handling patterns, ORM schemas, API routes, third-party SDK usage, and configuration. It writes a single file (COMPLIANCE_AUDIT.md) at the repository root with the audit report. No network calls, no shell commands, no environment variable reads, no telemetry. The "Network Access" permission auto-detected from the ZIP is a false positive — the skill discusses third-party APIs in its checklists but never invokes them.

    Listed6 months ago
    Updated24 days ago

    What's inside

    Frequently Asked Questions

    The skill analyzes your codebase to identify PII, mapping database models, API endpoints, and auth flows. It then matches these against specific legal requirements like GDPR, CCPA, and the EU AI Act to find vulnerabilities.

    The auditor is agent-agnostic and works with Claude Code, Cursor, VS Code Copilot, Gemini CLI, Windsurf, and Codex CLI. It is also stack-aware, recognizing popular ORMs like Prisma, Drizzle, and SQLAlchemy.

    The skill generates a 'COMPLIANCE_AUDIT.md' file that categorizes findings by severity (Critical to Low). Each entry includes the exact file and line number, the specific legal article cited, and a concrete technical fix.

    No manual configuration is required; the skill auto-detects your target markets using i18n files, domains, and currencies. It triggers automatically when you ask questions like 'Check my app for GDPR compliance.'

    This skill provides a technical engineering audit with specific code fixes, but it does not provide legal advice, generate Privacy Policy text, or offer official legal certification. It is designed to prepare you for production or formal reviews like SOC 2.

    The auditor includes 16 specialized checks for AI compliance, covering LLM disclosures, prompt injection risks, and EU AI Act Article 50 requirements regarding data sent to third-party providers.

    New in Security