Clapr REST API v1

A plain JSON API over the same data and permissions as the MCP server — built for n8n, Make, Zapier and your own scripts. Free on every plan.

Base URL: https://www.clapr.io/api/v1
Auth: Authorization: Bearer clapr_mcp_xxx — the same personal access token as MCP, generated in Settings → API & AI tools.
Format: JSON in, JSON out. Errors return { "error": "message" } with a matching HTTP status.

Quick start

curl https://www.clapr.io/api/v1/me \
  -H "Authorization: Bearer clapr_mcp_xxx"

Create a card:

curl -X POST https://www.clapr.io/api/v1/cards \
  -H "Authorization: Bearer clapr_mcp_xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "workspace_id": "…",
    "title": "Opus vs Gemini — real-world agent test",
    "description": "Hook: run the same 5 tasks on both, live.",
    "scheduled_date": "2026-07-14"
  }'

Endpoints

Method	Path	Description
`GET`	`/api/v1/me`	Your identity and workspaces — use this to test credentials.
`GET`	`/api/v1/workspaces`	All workspaces you can access, with your role.
`GET`	`/api/v1/workspaces/{id}/statuses`	Board columns of a workspace, in order.
`GET`	`/api/v1/cards`	List cards. Query: workspace_id, status_id, limit (max 200).
`POST`	`/api/v1/cards`	Create a card (title required; description, script, scheduled_date, status_id optional).
`GET`	`/api/v1/cards/{id}`	One card in full: script, references, comments, performance.
`PATCH`	`/api/v1/cards/{id}`	Update fields; set status_id to move the card to another column.
`POST`	`/api/v1/cards/{id}/comments`	Add a comment.
`POST`	`/api/v1/cards/{id}/references`	Attach a reference URL.
`GET`	`/api/v1/cards/{id}/performance`	Performance snapshots (views, likes, comments) over time.
`GET`	`/api/v1/openapi.json`	OpenAPI 3.1 spec (no auth required).

Using it from n8n, Make or Zapier

All three work with their generic HTTP node/module and header auth — no Clapr app needed. Step-by-step setups and ready-made recipes (auto-draft ideas, weekly performance report): Automation recipes.

Permissions, limits & errors

Writes require the owner or editor role in the target workspace; comments work for any member.
Rate limit: 120 requests/minute per token. Exceeding it returns 429 with a Retry-After header.
400 invalid input · 401 missing/invalid token · 403 insufficient role · 404 not found.
The API is for server-side use; browser (CORS) access is not supported.

Working from an AI coding agent instead? The MCP server gives it the same capabilities with zero HTTP plumbing.