Works with the AI Agents you already use
Basic Security: Pre-Launch Audit for AI-Coded Apps
Say "basic security" and your agent audits your AI-coded app against 70+ security checks and 17 privacy-law checks, then fixes what it safely can. Audit-only mode reports without changing code.
$5
Basic Security: Pre-Launch Audit for AI-Coded Apps
Example session with this skill installed
basic security audit only for my Next.js and Supabase app
- Read your context and instructions
- Compiled the basic security
I have completed the security audit of your Next.js and Supabase application. No files were modified.
Security Audit Report
| ID | Check | Status | Where | Recommendation |
|---|---|---|---|---|
| SEC-01 | Client-side secrets | Gap | src/lib/supabase.ts:4 | SUPABASE_SERVICE_ROLE_KEY is exposed in a frontend file. Move to server-side environment variables. |
| SEC-04 | Row-level security | Gap | sql/schema.sql | Table profiles has RLS enabled but uses USING (true). Change to auth.uid() = id. |
| SEC-11 | Rate limiting | Gap | src/app/api/chat/route.ts | No rate limit found on LLM endpoint. Add middleware to cap tokens per user. |
| SEC-24 | Security headers | Gap | next.config.js | Missing CSP and HSTS headers. Add standard browser security headers to config. |
| SEC-48 | Login settings | Needs you | Supabase Dashboard | Email confirmation is currently disabled. Enable this in Auth Settings to prevent fake signups. |
Next steps
- Run
basic securitywithout the "audit only" flag to automatically apply fixes for SEC-04, SEC-11, and SEC-24. - Manually move the service role key from
src/lib/supabase.tstoprocess.envon your hosting provider. - Visit the Supabase dashboard to enable email confirmation as noted in SEC-48.
Connects securely to your tools. The creator never sees your data.
About this skill
Built for apps made with AI coding tools (Cursor, Claude Code, Lovable, Bolt, Replit and similar), where the same security gaps show up again and again.
What it does
- SEC checks (SEC-01 to SEC-67, plus mobile SEC-M1 to M4): leaked API keys, missing Supabase/Firebase row-level security, broken auth and ownership checks, unsafe file paths, open redirects, webhook and payment flaws, LLM prompt-injection and cost-abuse risks, MCP/agent config risks, supply-chain and CI pipeline issues, and launch-readiness items such as backups, rate limits and account deletion.
- LAW checks (LAW-01 to LAW-17): privacy-law basics by region, including GDPR (EU), UK GDPR, CCPA/CPRA, COPPA, PIPEDA, Singapore PDPA, Malaysia PDPA, India DPDP and cross-border data transfers. Each says what can be fixed in code and what you must do yourself.
- Two modes: "basic security" finds and fixes gaps in code; "basic security audit only" reports every gap without changing anything.
- Plain-language report: a table of every check (Fixed / Already OK / Needs you / N/A), leading with what needs your action, explained in non-technical words.
- Every check is sourced: each one links to OWASP or the public research it came from.
- Optional weekly research: a prompt and scripts (works with any AI model) that look up new AI-coding security issues and list the ones the skill doesn't cover yet.
Disclaimer
This skill is provided as is, with no warranty. It is not legal, security, compliance or other professional advice. Automated checks and AI-made fixes can miss things or be wrong, and a clean result is not a guarantee that your app is secure or compliant. Review every change before you ship. If your app handles other people's data or money, have it checked by a qualified security professional or lawyer before launch.
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.
Cannot rotate leaked keys in provider dashboards. Legal text requires manual review. Git history rewrites require explicit user confirmation.
How to use
- 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
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 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 17 days ago
- Accounts
OpenAI
Google AI
Anthropic
Openrouter
- Software
Git
Permissions
Allowed Hosts