# KoboToolbox — how to use (mcp.ai)

Connect your KoboToolbox account and use 9 tools for forms and surveys straight from your AI agent. Connect with your own API key. KoboToolbox is a data collection platform for building forms, managing survey projects, collecting submissions, and exporting humanitarian and research data.

## Option A — via MCP (recommended)
Remote MCP endpoint (HTTP, streamable): `https://api.mcp.ai/p_kobotoolbox?ms=1787293440000`
Add it as a custom/remote MCP connector in your client (Claude, Cursor, VS Code…), then authenticate when prompted. Once connected, ask the agent to use the server's tools (e.g. `kobotoolbox_create_project`).

## Option B — via direct REST API
Base URL: `https://api.mcp.ai/api/kobotoolbox`
Auth: `Authorization: Bearer sk_live_…` — create a workspace API key at https://mcp.ai/settings/api-keys
Discover endpoints: `GET https://api.mcp.ai/api/kobotoolbox/_endpoints`

### Endpoints
- `POST https://api.mcp.ai/api/kobotoolbox/create/project` — Create a draft survey project, optionally by cloning an existing KoboToolbox asset.
  - body: { name: string, sector?: string, country?: string, clone_from?: string, description?: string, share_metadata?: boolean }
- `POST https://api.mcp.ai/api/kobotoolbox/delete/export` — Delete one asynchronous export job from a KoboToolbox project.
  - body: { export_uid: string, project_uid: string }
- `POST https://api.mcp.ai/api/kobotoolbox/delete/project` — Permanently delete one KoboToolbox project by UID.
  - body: { project_uid: string }
- `POST https://api.mcp.ai/api/kobotoolbox/deploy/project` — Deploy a draft KoboToolbox survey or redeploy its latest form version so it can accept and expose submissions.
  - body: { active?: boolean, project_uid: string }
- `POST https://api.mcp.ai/api/kobotoolbox/export/submissions` — Create an asynchronous KoboToolbox submission export and wait for it to complete, returning the export UID and result URL.
  - body: { query?: object, fields?: string[], flatten?: boolean, language?: string, export_type?: string, project_uid: string, submission_ids?: integer[], group_separator?: string, multiple_select?: string, include_media_url?: boolean, xls_types_as_text?: boolean, hierarchy_in_labels?: boolean, wait_timeout_seconds?: number, fields_from_all_versions?: boolean }
- `POST https://api.mcp.ai/api/kobotoolbox/get/project` — Retrieve a KoboToolbox project with its form content, settings, resource URLs, and deployment state.
  - body: { project_uid: string }
- `POST https://api.mcp.ai/api/kobotoolbox/get/submission` — Retrieve one submission by integer ID, root UUID, or submission UUID from a KoboToolbox project.
  - body: { project_uid: string, submission_id: string }
- `POST https://api.mcp.ai/api/kobotoolbox/list/assets` — Find projects and other visible assets in the connected KoboToolbox account, returning one page and a continuation cursor.
  - body: { limit?: integer, query?: string, ordering?: string, next_cursor?: string, current_user_permissions_only?: boolean }
- `POST https://api.mcp.ai/api/kobotoolbox/list/submissions` — Read one page of submissions for a deployed KoboToolbox project, with optional field selection, filtering, and sorting.
  - body: { sort?: string, limit?: integer, query?: string, fields?: string, next_cursor?: string, project_uid: string }

## Example prompts
- "What can I do in KoboToolbox?"
- "Show me a summary of my KoboToolbox account"

## More
- Page: https://mcp.ai/kobotoolbox
- Agent spec (llms.txt): https://mcp.ai/kobotoolbox/llms.txt
- Postman collection: https://mcp.ai/kobotoolbox/postman.json