io.github.mohammad-abbasi-sh/iskur-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| iskur_search_jobsA | Search İŞKUR's open job listings. keyword: Turkish job/occupation text (e.g. "garson", "aşçı") — empty means all jobs. city: province name (e.g. "İstanbul", "Ankara") or code (e.g. "34") — empty means all provinces. date_range: "all" (default), "last_24h", "last_week", "last_15d". work_type: "all" (default), "full" (Tam Zamanlı), "part" (Kısmi Zamanlı). education: "any" (default), "illiterate", "literate", "primary", "middle", "high", "associate", "bachelor", "master", "phd". page: results page (15 jobs per page; total and page_count are returned). limit: 1..15 jobs returned. Each job: job_no, title, sector, period, work_type, open_positions, location {scope, city, district}, deadline (dd.mm.yyyy), days_left, url. The search postback is slow (~40s); prefer page 1 and narrow filters. |
| iskur_latest_jobsA | Jobs posted on İŞKUR in the last 24 hours. city: province name (e.g. "İstanbul") or code (e.g. "34") — empty means all provinces. limit: 1..15 jobs returned. Each job: job_no, title, sector, period, work_type, open_positions, location {scope, city, district}, deadline, days_left, url. |
| iskur_get_jobA | Detail of one İŞKUR job listing. job_no: the listing number from a search result (e.g. "00009899349"). Returns description, required skills, occupation, experience (years/months), min/max education, and the detail URL. The detail page carries no deadline or city — those come from iskur_search_jobs — and no employer name (login-gated on the site). |
| iskur_list_citiesA | The 81 Turkish provinces İŞKUR searches by — code + name. Use the code (e.g. "34" for İstanbul) or the name in iskur_search_jobs / iskur_latest_jobs. Parsed live from the search form; falls back to the built-in list if the form cannot be read. |
| iskur_list_announcementsA | İŞKUR's official announcements (duyurular). limit: 1..10 announcements returned (the site lists ~10 per page). Each announcement: date, title, url. |
| iskur_healthA | Check whether İŞKUR's job portal is reachable right now. Returns status ok (HTTP 200), blocked (HTTP 403/429 — bot wall or rate limit) or error, plus the HTTP status and the probed URL. One polite request; it does not run a search. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 6 tools
Most tools are clearly separable: get_job (detail by job_no), list_cities (reference data), list_announcements, and health each have a distinct role. The one real overlap is iskur_latest_jobs versus iskur_search_jobs, since latest_jobs is effectively a pre-baked 'last 24h' search; the descriptions hint at this but do not explicitly tell the agent which to prefer. An agent could still pick correctly in most cases, but the boundary is soft.
All six tools share the iskur_ prefix and snake_case, which makes the set easy to scan. Most follow verb_noun (get_job, list_cities, list_announcements, search_jobs), but iskur_latest_jobs uses an adjective+noun form and iskur_health is a bare noun, so the pattern is not perfectly uniform.
Six tools is well-scoped for a single-source job board wrapper: search, latest, detail, reference list, announcements, and reachability check. Each tool earns its place and nothing feels like padding or a missing sibling.
The core lifecycle is covered: discover jobs (search/latest), paginate, drill into detail, resolve city codes, read announcements, and self-diagnose connectivity. Gaps are minor and partly imposed by the site (no employer name, no detail-page deadline), though the set offers no way to enumerate sectors, work types, or education levels as data the way it does cities.