security

Checks code changes against security and data-compliance rules during build, review, and QA.

1|Updated Jun 30, 2026
One-click install
npx skills add https://github.com/Lia-Creative/lia-plugins --skill security-lia-creative
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: security
Source: https://github.com/Lia-Creative/lia-plugins/tree/main/lia-tools/skills/security
Command: npx skills add https://github.com/Lia-Creative/lia-plugins --skill security-lia-creative

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Security issues are often discovered only at the end of a project, when fixing them is expensive. This Skill runs security and data-compliance checks continuously as work moves — at build prep, in the review loop, and at QA — so leaks of secrets, RLS violations, and personal-data mishandling are caught early. ## Core Features & Use Cases - Secret Leak Prevention: Enforces that no secret reaches a client bundle, that secrets live only in ~/.secrets/*.env, and that command-line arguments never expose tokens or passwords. - RLS and Auth Discipline: Requires migrations and policy review for changes to public.accounts, and mandates anon key + RLS over service keys on the client side. - Personal Data Minimization: Uses a schema map to flag personal and credential-adjacent fields so changes touching them get extra scrutiny. - Use Case: Before merging a pull request that adds a new environment variable to a Next.js app, run the check to confirm the variable does not carry NEXT_PUBLIC_, passes pnpm verify:bundle, and touches no signing path. ## Quick Start Ask the agent to run a security check on the current diff to confirm it does not leak secrets, weaken RLS policies, or expose personal data.

Frequently Asked Questions about security

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I check a code change for secret leaks before merging?▼

Run the security check at the review stage: verify no secret moves toward a client bundle, confirm server-only values go through a server-only module, and ensure no secret env var carries a NEXT_PUBLIC_ prefix. A bundle verification script in CI catches what review misses.

Where should secrets be stored in a development project?▼

Secrets should live in local files such as ~/.secrets/*.env, never in a repository, vault, ticket, or transcript. Reference them by location only, and if a credential leaks into a transcript or ticket, rotate it at the source and fix the leak path.

When should I use the anon key versus the service role key in Supabase?▼

Use the anon key with Row Level Security for anything client-side; the service role key bypasses RLS and must never reach client code. Changes to tables like public.accounts should go through migrations with policy review first.

Why do command-line arguments leak credentials?▼

Process argument lists are world-readable via tools like ps, so flags such as --password or --token expose values to any user on the machine. Redact such arguments before printing command lines and treat any leaked value as a compromised credential.

What are the limitations of a rule-based security checklist?▼

A checklist only covers cases it has explicit rules for, so it cannot catch novel vulnerability classes or replace a full audit. The mitigation is to name what the check could not cover on each run and add new rules only when a real incident or finding pays for them.