ACP sits between your AI client and the tools it calls. Install the hook (or point at our MCP endpoint) and every tool call gets identity, policy, and audit — without changing how your team works.
Building the agent yourself (CrewAI, LangGraph, Anthropic Agent SDK, OpenAI Agents SDK)? See Frameworks for guided starter code.
Step-by-step install guides for each AI client — with the caveats vendor docs leave out. Two install paths: a local hook (for coding agents) or an MCP server connection (for chat clients).
Using a client we don't list? If it speaks MCP (Windsurf, Continue, ChatGPT’s connectors) or the OpenAI chat-completions API, point it at ACP — same governance, no code change.
Want cost too? Governance hooks cover tool calls; to meter model spend, cache-hit rate, and the loop-vs-leaf cost X-ray, route your model calls through the proxy — one base-URL change.
ACP verifies JWTs against your IdP's JWKS endpoint on every call. Any OIDC-compliant provider works out of the box — your users stay in your directory.
Regardless of which client you install — every agent action passes through the same governance pipeline.
Free to start. Add governance to your existing AI client in minutes.