Local control plane for AI coding agents — trust-gate, firewall, and audit every MCP tool call. One manifest for servers, skills, instructions, and settings across Claude Code, Codex, Cursor, Gemini CLI + 9 more.
<img alt="agentstack" src="docs/logo.svg" width="380"> > **Cloning a repo shouldn't hand your agent to a stranger.** > AgentStack is a local control plane for AI coding agents: one reviewed > manifest defines what every agent CLI may run — then trust-gates, > firewalls, and audits it. Nothing runs until it's trusted, and nothing