AgentShield is a security‑focused scanner for Claude Code AI agent configurations. It analyzes `.claude/` files for hard‑coded secrets, overly permissive tool permissions, unsafe hooks, risky MCP servers, and prompt‑injection vectors, and can be run as a CLI, GitHub Action, ECC plugin, or GitHub App (with optional auto‑fixing and deep Opus 4.6 multi‑agent analysis). The tool is aimed at developers and teams building or deploying AI agents who need to harden their agent setups before going live.
How the donated funds are distributed
Kivach works on the Obyte network, and therefore you can track all donations.