> For the complete documentation index, see [llms.txt](https://cerebral-systems.gitbook.io/cerebral-systems-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://cerebral-systems.gitbook.io/cerebral-systems-docs/index.md).

# Index

**A local policy and execution boundary for AI coding agents.**

> **Current release:** Warrant v0.3.0 ships as a local macOS CLI through Homebrew. It requires no Warrant account, API key, or hosted endpoint. The hosted dashboard and HTTP API are not publicly released.

### Start here

```sh
brew install cerebral-systems/tap/warrant
warrant --version

cd /absolute/path/to/workspace
warrant setup claude --policy read-only
```

Follow setup's printed `Next:` instruction, return to the same workspace, then run:

```sh
warrant doctor claude
claude
```

Follow the complete local CLI quickstart.

### What ships in v0.3.0

| Surface                           | Current status                                                                        |
| --------------------------------- | ------------------------------------------------------------------------------------- |
| Distribution                      | Apple Silicon and Intel macOS binaries through Homebrew.                              |
| Claude Code                       | Primary demo-ready, version-pinned local integration.                                 |
| Codex, Hermes, OpenClaw, PicoClaw | Experimental exact-version adapters; clean runtimes do not receive model credentials. |
| Local policies                    | Five reviewed templates plus editable, closed JSON v2 policy files.                   |
| Local operations                  | Bounded workspace list, read, search, exact write, and automation-file write.         |
| Approvals                         | Hidden local-owner authority over exact supported writes.                             |
| Assurance                         | `integrated`; not OS-contained and not `strict_verified`.                             |

The released templates are `read-only`, `owner-approved`, `docs-only`, `workspace-development`, and `automation-locked`. Current source adds a sixth, `privacy-locked`, for the next release. Use `warrant policies list` and `warrant policies show <template>` to inspect them, then follow Local policies to create, validate, diff, and activate a custom policy.

### Next-release source preview

The source tree includes a version-pinned Grok Build adapter and the narrow `privacy-locked` read policy:

```sh
warrant setup grok --policy privacy-locked
grok
```

The managed Grok process starts in an empty Warrant session workspace; Warrant exposes the original workspace through its brokered MCP operations. This is unreleased and does not prevent an unmanaged same-user process from using ambient filesystem or network access. Follow the Grok Build preview flow before testing it.

### Boundary

```
agent → managed shim / hook / MCP → local policy → approval → workspace actuator
```

Warrant pins its own runtime, the agent launcher and interpreter, the workspace policy, and the exact supported operation. It fails closed when that evidence drifts.

It does not stop a separate same-user process that retains ambient filesystem, shell, credential, or raw-network access. It does not locally actuate Kubernetes, Stripe, Postgres, arbitrary commands, deletes, renames, package installs, or network requests.

### Documentation paths

| Goal                                | Start here                   |
| ----------------------------------- | ---------------------------- |
| Install and protect a workspace     | Local CLI quickstart         |
| Understand or switch policies       | Local policies               |
| Check agent-specific support        | Agent integrations           |
| Understand guarantees and gaps      | Security architecture        |
| Check planned work                  | Product status and roadmap   |
| Read server implementation material | Unreleased server/API design |

The server/API pages are retained for development and future self-hosting work. They do not describe a supported public endpoint or current onboarding path.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://cerebral-systems.gitbook.io/cerebral-systems-docs/index.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.