Finds hiring companies, enriches their data, and finds lead emails for sales prospecting
Server has 4 tools with basic descriptions and input schemas, but significant gaps in parameter documentation, output schema clarity, and error handling guidance. Tool names are action-oriented (search_jobs, enrich_company, find_lead_email, generate_sales_leads), which is good. However, parameter descriptions are minimal or missing type info. Output schemas are not formally documented, responses are inferred from code. Error handling returns generic dicts without recovery guidance. No tool annotations (readOnlyHint, etc.). Descriptions average ~100 chars, below the 194-char baseline for A+ tools. Parameters lack constraints (enums, ranges, formats). No pagination support despite potential for large result sets.
Enriches a company domain with detailed data using TechnologyChecker API. Returns industry, size, technology stack, etc.
Finds a decision-maker's email and name at a given company domain using Hunter.io. Uses domain search to find the most relevant lead.
Master pipeline: Finds hiring companies, enriches their data, and finds lead emails. Returns a complete list of qualified sales leads with contact details, or an error dictionary.
Searches for jobs using JSearch API (via RapidAPI). Returns a list of companies that are hiring or an error dictionary.
Output schemas not documented. Tool responses are inferred from code (dicts with varying fields). LLMs cannot plan downstream calls or extract required fields reliably.
Parameter descriptions are minimal or missing. 'num_pages' in search_jobs has no type hint in description (is it int or string?). 'domain' and 'company_name' in find_lead_email lack format/constraint guidance.
Error responses are generic dicts without recovery guidance. E.g., 'Company enrichment failed: <error>' tells LLM nothing about retry eligibility or next steps. No error classification (retryable vs fatal).
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 47 | 2026-07-28+ | v2 |
No pagination support. search_jobs and generate_sales_leads return lists but lack limit/offset params or total_count in response. Hardcoded limit of 5 in generate_sales_leads is opaque to LLM.
No tool annotations. Tools lack readOnlyHint (all are read-only), idempotentHint, or destructiveHint. LLMs cannot infer safety properties without explicit hints.