Swiss labor market data server. Liefert die nationale Jahresreihe der registrierten Arbeitslosen und Stellensuchenden (SECO-Zahlen, publiziert vom BFS) sowie die UVG-Unfallstatistik. Monats- und Kantonswerte gibt es zurzeit in keiner maschinenlesbaren Quelle; die betroffenen Werkzeuge sagen das, statt eine nationale Zahl kantonal zu beschriften. All data is public and requires no API key. Particularly useful for educational planning, vocational guidance (Berufswahlberatung), and apprenticeship market monitoring (Lehrstellen-Monitoring).
Strong foundation with 10 well-named tools using consistent verb_noun patterns (search_, get_, unemployment_, job_). All tools have descriptions (avg 120 chars, within 10-1024 range). Input schemas are present and properly typed with JSON Schema. However, output schemas are not documented, LLMs cannot predict response structure. Parameter descriptions are adequate but lack actionable constraints (e.g., no examples of valid canton codes in descriptions, only in schema). Error handling and recovery guidance are absent. No tool annotations (readOnlyHint, etc.) despite all being read-only operations.
Get job seeker (Stellensuchende) statistics. Returns data on registered job seekers, optionally filtered by canton.
Get the URL of a monthly press report (Monatsbericht) from arbeit.swiss. Returns the URL for a specific month and language.
Get open positions (Offene Stellen) statistics. Returns data on registered job openings.
Search for SECO datasets on opendata.swiss. Searches for datasets related to Swiss labor market data, unemployment statistics, job openings, and occupational data.
Get unemployment statistics by occupation (Berufshauptgruppe). Returns data on registered unemployed broken down by occupational groups.
Get unemployment statistics (registered unemployed and job seekers) from SECO/BFS. Returns national annual time series data, optionally filtered by canton and year.
Get UVG (occupational accident insurance) statistics by economic branch (NOGA). Returns accident and insurance data broken down by industry sector.
Output schemas not documented. LLMs cannot predict response structure, field names, or types. This forces agents to guess at downstream field mappings and risks parsing errors.
No tool annotations (readOnlyHint, destructiveHint, idempotentHint). All 10 tools are read-only but lack explicit annotation. Agents cannot distinguish safe tools from destructive ones without annotations.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 57 | 2026-07-28+ | v2 |
Get UVG (occupational accident insurance) overview statistics. Returns summary data on occupational accidents and insurance statistics.
Get UVG (occupational accident insurance) trend analysis. Returns time series data on occupational accident trends and statistics.
Get youth unemployment statistics for 15–24 year olds. Returns data on registered unemployed youth, optionally filtered by canton.
Parameter descriptions lack actionable constraints. Canton codes documented as '2-letter' but no examples (ZH, BE, GE) in descriptions, only in schema. Year ranges and response_format enums are in schema but not reinforced in descriptions for LLM clarity.
No error handling or recovery guidance. Tools do not document what happens on invalid canton codes, network failures, or missing data. LLMs receive no guidance on retry strategy or fallback actions.
Descriptions for open_positions, uvg_overview, uvg_trends are generic (65 chars). They lack context on when to use each tool vs alternatives or what data structure is returned.