# Rubric

## Docs

- [Claude Agent adapter](https://docs.rubric-app.com/adapters/claude-agent.md): Install a `PreToolUse` hook so Claude Agent's tool calls run through governance.
- [Claude Code adapter](https://docs.rubric-app.com/adapters/claude-code.md): Gate every Claude Code tool call through your Rubric policies. <30ms p99 per call.
- [LangChain adapter](https://docs.rubric-app.com/adapters/langchain.md): Wrap any list of LangChain tools so each invocation is governed.
- [MCP adapter](https://docs.rubric-app.com/adapters/mcp.md): Govern any MCP-speaking agent with one wrapper call.
- [Changelog](https://docs.rubric-app.com/changelog/overview.md): New features, updates, and fixes shipped to Rubric.
- [How it works](https://docs.rubric-app.com/concepts/architecture.md): What the Rubric SDK does in your process and what it sends over the wire.
- [DLP](https://docs.rubric-app.com/concepts/dlp.md): Inline detection of secrets, PII, and PHI in tool arguments.
- [Evaluation](https://docs.rubric-app.com/concepts/evaluation.md): What happens when you call `gov.evaluate(...)`.
- [Identities & enrollment](https://docs.rubric-app.com/concepts/identities.md): Per-agent signed identities, enrollment tokens, and how the SDK self-bootstraps.
- [Policies](https://docs.rubric-app.com/concepts/policies.md): YAML in the dashboard, versioned, diffable, dry-runnable.
- [Traces](https://docs.rubric-app.com/concepts/traces.md): Attach the conversation transcript to every audit event.
- [Agents](https://docs.rubric-app.com/dashboard/agents.md): Manage the fleet — view, register, freeze, assign owners.
- [Audit log](https://docs.rubric-app.com/dashboard/audit-log.md): Every governed tool call, filterable, with traces.
- [Enrollment](https://docs.rubric-app.com/dashboard/enrollment.md): Create, revoke, and renew enrollment tokens.
- [Incidents](https://docs.rubric-app.com/dashboard/incidents.md): Auto-opened on deny spikes and critical DLP. Triage in seconds.
- [Dashboard overview](https://docs.rubric-app.com/dashboard/overview.md): What each section of the Rubric dashboard is for.
- [Policies](https://docs.rubric-app.com/dashboard/policies.md): Author, version, dry-run, publish.
- [Installation](https://docs.rubric-app.com/guides/installation.md): Install the SDK in your agent project and point it at your Rubric API.
- [Quickstart](https://docs.rubric-app.com/guides/quickstart.md): Install the SDK, enroll an agent, ship your first governed tool call. ~5 minutes.
- [Rubric — runtime governance for AI agents](https://docs.rubric-app.com/index.md): Per-agent identities, runtime policy enforcement, and an audit trail for every tool call your agents make.
- [Core building blocks](https://docs.rubric-app.com/sdk-node/core.md): The four primitives exported from @rubric-app/core.
- [Errors](https://docs.rubric-app.com/sdk-node/errors.md): Exception classes exported by @rubric-app/core.
- [Evaluator](https://docs.rubric-app.com/sdk-node/evaluator.md): Synchronous, in-process policy evaluator.
- [Install](https://docs.rubric-app.com/sdk-node/install.md): Install @rubric-app/core or @rubric-app/claude-code.
- [DLP](https://docs.rubric-app.com/sdk-python/dlp.md): Detector configuration and the detection result shape.
- [Errors](https://docs.rubric-app.com/sdk-python/errors.md): Exception types raised by the SDK.
- [evaluate()](https://docs.rubric-app.com/sdk-python/evaluate.md): The hot-path function. Evaluate a tool call against the bundle and ship an audit event.
- [Governance](https://docs.rubric-app.com/sdk-python/governance.md): The underlying class. Use the decorator API for most code.
- [init() & @tool](https://docs.rubric-app.com/sdk-python/init.md): The top-level decorator API. Boot once, decorate functions, ship.
- [TraceContext](https://docs.rubric-app.com/sdk-python/trace.md): Build the conversation transcript that gets attached to audit events.

## OpenAPI Specs

- [openapi](https://docs.rubric-app.com/api-reference/openapi.json)