> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.cloudraker.com/paperwork/developers/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.cloudraker.com/_mcp/server.

# Developers

## Docs

- [Developer guide](https://docs.cloudraker.com/paperwork/developers/overview.md): Build on CloudRaker. Authenticate, process documents, and automate your workflows over the API.
- [Quickstart](https://docs.cloudraker.com/paperwork/developers/quickstart.md): Get a key. Extract structured data from a PDF in one call. Then scale it up.
- [Authentication](https://docs.cloudraker.com/paperwork/developers/authentication.md): Authenticate every request with a bearer token: an organization API key or a session JWT.
- [Files](https://docs.cloudraker.com/paperwork/developers/files.md): Register a document once, by URL or presigned upload. Then reuse it across runs.
- [File tools](https://docs.cloudraker.com/paperwork/developers/file-tools.md): Convert, render, split, join, and redline registered files. No model is involved.
- [Spaces](https://docs.cloudraker.com/paperwork/developers/spaces.md): A named container for files and runs, with a lifetime you set. The whole API is mirrored inside it.
- [Runs](https://docs.cloudraker.com/paperwork/developers/runs.md): One id, one lifecycle. Track it, download what it produced, keep it or let it expire.
- [Agent runs](https://docs.cloudraker.com/paperwork/developers/agent-runs.md): Start an agent on a set of files. Then answer its sign-offs and human steps over the API.
- [Webhooks](https://docs.cloudraker.com/paperwork/developers/webhooks.md): Receive signed events when runs progress. Verify them with public-key JWTs.
- [Process API](https://docs.cloudraker.com/paperwork/developers/process-api.md): Upload documents and run actions in one call. Then poll for parsed text, extractions, and evidence.
- [Files and uploads](https://docs.cloudraker.com/paperwork/developers/files-and-uploads.md): Put files into a space with a presigned upload. Then poll until parsing finishes.
- [SDKs](https://docs.cloudraker.com/paperwork/developers/sdks.md): Typed clients for TypeScript and Python that handle the base URL and bearer token for you.
- [CLI](https://docs.cloudraker.com/paperwork/developers/cli.md): The `paperwork` command line: every API endpoint as a subcommand, with browser sign-in or API-key auth.
- [MCP server](https://docs.cloudraker.com/paperwork/developers/mcp.md): Drive the whole CloudRaker API from Claude and other MCP clients through Code Mode.
- [OpenAI-compatible API](https://docs.cloudraker.com/paperwork/developers/openai-compatibility.md): Point any OpenAI client at Paperwork. Parse, extract, and redact documents through /chat/completions.
- [Agent quickstart](https://docs.cloudraker.com/paperwork/developers/agent-quickstart.md): For coding agents: fetch the contract, run one call, then build.
- [API context for agents](https://docs.cloudraker.com/paperwork/developers/agents.md): The whole CloudRaker API contract on one page, written for a coding agent to fetch as Markdown.
- [Versioning and compatibility](https://docs.cloudraker.com/paperwork/developers/versioning.md): What we can change without warning, and what your integration must tolerate.
- [Errors](https://docs.cloudraker.com/paperwork/developers/errors.md): How the API reports failures: status codes and the error body shape.
- [Rate limits](https://docs.cloudraker.com/paperwork/developers/rate-limits.md): The request budget of the capability API, the shape of a 429, and how to back off.
- [Pagination and filtering](https://docs.cloudraker.com/paperwork/developers/pagination-filtering.md): How listing works per endpoint group, and how to filter and sort data objects.