Developer documentation
Everything you need to build it secure.
Policy on every action your AI agents take: guides, references and recipes for intent enforcement and agent governance.
Start here
Govern in your runtime
- Intent GatewayGovern an agent in five minutesChange one base URL. Every tool call your agent's model returns is checked against your policy before your code runs it.
- ToolsGuard your coding assistantClaude Code, Codex, Copilot and Gemini under one policy, with a console that shows what each one did.
- SDKBuild enforcement into your codeCapture the plan, verify the intent token and check each step with the Python or TypeScript SDK.
- Intent ContainerHold a sandboxed agent to its objectiveEnforce each task’s authority on files, processes and network calls, from the request path down to the kernel.
Browse
All documentation
- Need a handSearch the docs with ⌘ K, or ask the team in Discord.