Coding agent documentation index Fetch the complete documentation index at: https://docs.worklittle.com/docs-agent-manifest.json Use this file to discover all available pages before exploring further.

Rate limits

Per-minute throttles and free monthly quotas for job search, apply, and MCP job tools.

API keys

Default 60 requests per minute per key (1-minute KV window). Response: HTTP 429, error.code = RATE_LIMITED, Retry-After: 60.

Implements exponential backoff in clients and agents that call GET /jobs, GET /companies, apply routes, and document tools.

Free monthly quotas (UTC)

| Product | Endpoint or trigger | Free quota (UTC) |
| --- | --- | --- |
| **Job search** | `GET /jobs` / MCP `search_jobs` | **1000** jobs returned / month |
| **Job detail (raw / cached AI)** | `GET /jobs/:id` / MCP `get_job_details` | **Free** - raw `description_text` and already-cached AI never bill |
| **Job detail AI generation** | `GET /jobs/:id?summary=true` first-time only | **100** free generations/org/**day**, then token rates. Not a hard deny |
| **Job salary average** | `GET /jobs/salary-average` | **1000** requests / month |
| **Company map** | `GET /jobs/map` | **1000** requests / month |
| **Company search** | `GET /companies` | **1000** companies returned / month |

When a hard quota is exceeded, the API returns QUOTA_EXCEEDED (HTTP 429). Inquire at [business@worklittle.com](mailto:business@worklittle.com) for higher limits.

MCP job tools

MCP forwards Jobs API rate limits for tools like search_jobs, get_job_details, and apply helpers. A 429 from api.worklittle.com surfaces as a JSON-RPC error on the MCP client. Retry after backoff.