Skip to Content
The Makinari API repo now includes an MCP Server — connect your AI models directly. View on GitHub →
API referenceRequirements

Requirements

AI Goals. Tools: requirements (CRUD) and HTTP GET /api/requirements for a site-scoped list.

POST/api/agents/tools/requirements/create · /get · /update
ActionRouteRequired
createPOST /api/agents/tools/requirements/createsite_id, title
listPOST /api/agents/tools/requirements/getsite_id
updatePOST /api/agents/tools/requirements/updatesite_id, requirement_id

status: backlog, validated, in-progress, on-review, done, canceled.
completion_status: pending, completed, rejected.
priority: high, medium, low.

instructions should say whether work is an external deliverable or uses Makinari tools. budget is the token cap.

curl -X POST "https://backend.makinari.com/api/agents/tools/requirements/create" \ -H "Authorization: Bearer YOUR_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "site_id": "YOUR_SITE_ID", "title": "Publish a sales blog post", "type": "content", "priority": "high", "status": "backlog", "instructions": "Write and publish using Makinari tools. Target VP Sales." }'

Status list (dashboard-style):

curl -X GET "https://backend.makinari.com/api/requirements?site_id=YOUR_SITE_ID" \ -H "Authorization: Bearer YOUR_API_KEY"

MCP: Platform ops. Concepts: Requirements vs workflows.

Last updated on