MCP server for job post verification and scam detection with web evidence gathering tools
HireProof MCP server has 4 tools with basic schemas and descriptions, but significant gaps in parameter documentation and output schema clarity. All tools have descriptions (10-20 chars baseline met), but lack depth for LLM decision-making. Parameter descriptions are minimal or absent. No error handling guidance, no output schemas documented, no tool annotations. Naming is reasonable (verb_noun pattern: search_*, news_*, jobs_*, local_*) but descriptions are too brief to guide LLM selection effectively. No evidence of pagination, result limits, or chaining IDs in responses.
Find comparable job listings from legitimate companies to benchmark salary, role, and requirements
Check for local business footprint, maps, directories, business registration, and location signals
Search for recent news, reputation signals, scam reports, and media mentions about a company or role
Search for company web presence, including official website, LinkedIn profile, domain registration, and business information
Output schemas not documented. No visibility into what fields tools return, forcing LLMs to guess downstream field names and breaking tool chaining.
Descriptions too brief (35-45 chars). Do not explain WHEN to use each tool or how they differ. LLMs cannot distinguish search_company from local_presence without longer, contextual descriptions.
Parameter descriptions missing or minimal. 'keywords' in news_check lacks guidance on format/count. 'level' in jobs_compare has no enum or examples. LLMs will hallucinate invalid values.
No error handling guidance. Tools return no recovery hints if company not found, news unavailable, or location invalid. Agents cannot self-correct.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 47 | 2026-07-28+ | v2 |
No pagination or result limits documented. If jobs_compare returns 500 listings, context window explodes. No limit parameter visible in schema.