Jamaalx

LLM Security — Security skill for Claude Code

Security community

AI/LLM agent security audit (OWASP GenAI/LLM Top 10 2026): prompt injection (direct + indirect/RAG), jailbreaks, system-prompt & secret leakage, excessive agency / tool abuse, RAG multi-tenant exfiltr.

How to install LLM Security

Installs to ~/.claude/skills/jamaalx-claude-skills-llm-security/SKILL.md

Terminal
mkdir -p ~/.claude/skills/jamaalx-claude-skills-llm-security && curl -fsSL https://raw.githubusercontent.com/Jamaalx/claude-skills/HEAD/commands/llm-security.md -o ~/.claude/skills/jamaalx-claude-skills-llm-security/SKILL.md

Restart Claude Code, or start a new session, for it to be picked up.

What LLM Security does

description: "AI/LLM agent security audit (OWASP GenAI/LLM Top 10 2026): prompt injection (direct + indirect/RAG), jailbreaks, system-prompt & secret leakage, excessive agency / tool abuse, RAG multi-tenant exfiltration, insecure output handling, cost-DoS, model supply-chain. Includes a red-team probe set. Generates LLM-SECURITY-FIXES.md. For AI assistants, chatbots, support agents, and agentic RAG systems." allowed-tools: [Bash, Read, Glob, Grep, Agent, WebSearch, WebFetch, TaskCreate, Task

Alternatives in Security

  • Spec Check — Audit GWT acceptance test specs for implementation leakage 149 ★
  • Sovereign Skills — 20 production-grade skills for AI coding agents — setup, scope, discipline, code review, security, session man 125 ★
  • Security Audit Stride — Chạy checklist bảo mật OWASP Top 10 + STRIDE trước bước review cuối của Tech Lead trong WF-REVIEW-CRIT (luôn c 90 ★

Full documentation available on GitHub

View Source Repository

Related Skills

API Security — Security skill for Claude Code

API Security

API security audit (OWASP API Security Top 10 2023): BOLA/IDOR, broken authentication, broken object-property-

Security community
Rls Audit — Security skill for Claude Code

Rls Audit

Supabase RLS & Postgres authorization audit: unprotected tables, permissive policies, self-elevation vectors,

Security community
Agent Name Service — Security skill for Claude Code

Agent Name Service

Agent Name Service (ANS) Protocol, introduced by the OWASP GenAI Security Project, is a foundational framework

Security community
Authz — Security skill for Claude Code

Authz

Hunt broken access control — IDOR, missing role checks, mass assignment, and tenant isolation gaps. OWASP A01,

Security community
Security Auditor — Security skill for Claude Code

Security Auditor

Audits code for OWASP-class vulnerabilities — injection, auth flaws, secret leaks, unsafe deserialization.

Security community
Security Audit — Security skill for Claude Code

Security Audit

Audit de securite complet d'une web app (OWASP Top 10, CWE/CVE, headers, auth, paywall, infra). Genere un rapp

Security community

Related Agents

Kavach LLM — Security agent for Claude Code

Kavach LLM

KAVACH LLM/AI security specialist. Audits prompt injection (direct + indirect/RAG), jailbreak/persona/system-p

Security community
Ehs AI Safety — Security agent for Claude Code

Ehs AI Safety

AI, agent and chatbot security specialist for the Ethical Hacker Squad. Reviews the instruction/data boundary,

Security community
LLM Sec Review — Security agent for Claude Code

LLM Sec Review

LLM/agent security review specialist — prompt injection, the Agents Rule of Two, tool-call authorization, mode

Security community