MCP server providing tools to search and discover public APIs from the ls-apis dataset. Allows filtering by query, category, authentication type, and listing API categories and data providers.
Three tools with clear verb-noun naming (search_apis, list_categories, list_providers) and concise descriptions (50-80 chars). All tools have input schemas with type definitions. However, parameter descriptions are minimal or absent: search_apis has 4 params with basic descriptions but lacks constraints (enums for auth, min/max for limit); list_categories and list_providers have empty input schemas with no documented output structure. No error handling guidance, no pagination support despite search_apis returning potentially large result sets, and output schemas are not formally documented. Responses are JSON-stringified but lack field-level documentation for downstream tool chaining.
List all API categories with the count of APIs in each, sorted by popularity.
List all data providers with the number of APIs contributed by each, sorted by popularity.
Search public APIs from the ls-apis dataset. Filter by query, category, auth type, and limit results.
Output schemas not documented. Tools return JSON but LLMs cannot infer field names, types, or structure for downstream chaining. search_apis returns {total, results} but result object fields are undocumented.
search_apis lacks pagination. No limit default, no offset/page params, no next_cursor. Large result sets will blow context windows. Baseline: paginated tools accept limit (1-100) and offset/cursor.
Parameter 'auth' in search_apis lacks enum constraint. Description says 'e.g. apiKey, OAuth, no' but does not declare valid values formally. LLMs will hallucinate invalid auth types.
Parameter 'limit' in search_apis has no min/max bounds. Unbounded numeric params let LLMs pass absurd values (e.g. limit=999999) that waste resources.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 52 | 2026-07-28+ | v2 |
No error handling guidance. Tools do not document what happens on invalid query, missing category, or data load failure. LLMs receive no recovery hints.