Baton Manual

User manual

Baton, every screen step by step

How to send requests and read what came back, keep variables and secrets apart, set up auth and scripts, work with GraphQL and WebSocket, bring collections in from other clients, sync through git, run a worker, and let an AI agent help — with an annotated screenshot of each screen.

Start readingDownload the PDFTry it in the browser
  1. Chapter 1About BatonWhat Baton is: an API client whose collections, environments and schedules are plain YAML files in a git repository you own.
  2. Chapter 2The windowFind your way around Baton: the sidebar, the request editor, the response panel, the status rows and the command palette.
  3. Chapter 3WorkspacesOpen, create and switch Baton workspaces — folders of plain YAML files that are git repositories you own.
  4. Chapter 4Collections and requestsCreate collections, import from Postman, OpenAPI, Bruno, Insomnia, Yaak or curl, follow an OpenAPI spec, and export back out.
  5. Chapter 5Sending a requestSend a request in Baton and read what came back: the response, what was really sent, the console, history and saved examples.
  6. Chapter 6Variables and environmentsUse {{variables}} in Baton, switch environments, and mark one as production so an agent's send into it asks you first.
  7. Chapter 7SecretsHow Baton keeps secret values out of your repository: names in the file, values in the OS secure store or HashiCorp Vault.
  8. Chapter 8AuthorizationBearer, basic, API key, AWS Signature, digest and OAuth 2.0 in Baton — set once on a collection, inherited by its requests.
  9. Chapter 9Request bodiesJSON, text, form, multipart, binary and GraphQL request bodies in Baton, and how each is sent.
  10. Chapter 10Scripts and testsPre-request and test scripts in JavaScript with Baton's baton.* API, sandboxed the same way in the app and on a schedule.
  11. Chapter 11GraphQL and WebSocketGraphQL in Baton with schema introspection and a query builder, and WebSocket sessions with a live message log.
  12. Chapter 12Running a collectionRun every request in a Baton collection in order and read the results, request by request and test by test.
  13. Chapter 13Git sync and the workerCommit, push and pull from Baton; run collections on a schedule, in CI, and as a mock server on a worker you host.
  14. Chapter 14AI agentsConnect Claude Code, Cursor, VS Code or Codex to Baton over MCP: edits arrive as drafts you accept, and production asks first.
  15. Chapter 15Settings, licensing and troubleBaton's settings, the free license key and how it is checked offline, and what the common error messages mean.
  16. Chapter 16The desktop buildWhat the Baton desktop app adds on macOS, Windows and Linux: real folders, the system git, cookies, and updates.