Agent Skill · Worklittle
job-search
Guides Worklittle job search and discovery: search_jobs / GET /jobs filters, quota-aware pagination, get_job_details vs get_job_keywords, closed jobs, and free market stats. Use when searching jobs, building a job board or search assistant, ranking roles, or answering market questions from Worklittle data.
Skill body
Canonical funnel
search_jobs / GET /jobs (limit ≤ 20)
→ rank / pick with the user
→ get_job_details / GET /jobs/:id # shortlisted only
→ get_job_keywords # for resume/cover tailoring
Do not feed full job descriptions into document models — use keywords. See references/filters.md.
Critical rules
- Search is free within monthly quotas (not billed per row). Still keep
limitat 10–20 so you do not burn quota. - Matching is title/filter text (substring / FTS), not vector embeddings.
- List rows are snippets. Never state requirements or salary from search results alone.
- Opaque cursors only. Max 50 per page. Paginate on user intent, not by default (~3 pages/conversation).
- Closed jobs: check
closed_at. Open ≠ applicable. - Aggregates are free:
GET /stats/get_market_overviewfor counts and trends — do not page search to count.
Endpoints and tools
| Goal | REST | MCP | Scope |
|---|---|---|---|
| Search | GET /jobs |
search_jobs |
jobs:read |
| Detail | GET /jobs/:id |
get_job_details |
jobs:read |
| Keywords | — | get_job_keywords |
jobs:read |
| Market stats | GET /stats |
get_market_overview |
jobs:read |
| Map | GET /jobs/map |
— | jobs:read |
Docs: Search jobs · Job search assistant · Pagination
Work with this as data
Every skill here is available over the APIs.io API and to AI agents over MCP.