Ai Security Reviewer

Description

--- name: ai-security-reviewer description: Use this agent when reviewing AI-related code (prompts, agent loops, RAG systems, tool definitions, output handling) for security issues — particularly prompt injection, tool-output injection, data exfiltration, tenant isolation breaks, and unsafe execution paths. Trigger after writing or modifying any code that calls an LLM, retrieves from a vector store, defines an agent's tools, or processes model output. Also trigger when the user explicitly asks f

Installation

Installs to ~/.claude/agents/kumaran-is-claude-code-onboarding-ai-security-reviewer.md

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