JI LABS

Developers

A clear, versioned starting point for agents and developers who need to begin a Ji Labs conversation programmatically.

When to use Ji Labs

Use this API when an agent or internal tool needs to start a conversation about automating procurement, supplier research, document-heavy workflows, repetitive back-office work, or decisions built from scattered business context. Ji Labs is not a general-purpose AI model, search index, or data-export service.

Quickstart

The inquiry endpoint accepts JSON at the stable /api/v1/inquiry path.

curl https://jilabs-ai.com/api/v1/inquiry \
  -H 'Content-Type: application/json' \
  -d '{"name":"Ada Example","companyName":"Example Co","email":"ada@example.com","inquiry":"We want to explore supplier research automation."}'

Sandbox-safe smoke test: GET /api/v1/health only reports service availability; it does not create or modify data.

Authentication, privacy, and limits

The public inquiry endpoint does not require an API key. API responses include RateLimit-Limit, RateLimit-Remaining, and RateLimit-Reset so clients can self-throttle. A throttled request returns HTTP 429 with Retry-After. Do not send credentials, secrets, or unnecessary personal data. Review the privacy notice before integrating.

Versioning and errors

Integrate with /api/v1/ paths. Version 1 has stable operation IDs in the OpenAPI document. The old unversioned inquiry path remains for the existing website form but is deprecated; its responses include Deprecation, Sunset, and a successor Link header. Errors are JSON objects with error.code, error.message, and error.resolution.

CLI status

Ji Labs does not publish an official CLI yet. The public surface is intentionally small, so curl or any HTTP client is the supported automation workflow until a CLI would provide meaningful additional value.