MCP server for searching Google Scholar — papers, authors, citations, and BibTeX
Server has 6 well-named tools with complete input schemas and descriptions. All tools follow verb_noun naming (search_papers, get_paper_details, etc.). Descriptions are present and substantive (100-200 chars typical). Input schemas include types, defaults, and constraints. However, output schemas are not documented in the source, only Pydantic model classes are referenced without inline schema documentation. Error handling is minimal: bulk_search catches exceptions but provides generic error messages. No tool annotations (readOnlyHint, destructiveHint) despite all tools being read-only. Parameter descriptions could be more explicit about constraints (e.g., max_results 1-20 is stated but not enforced in schema). Missing guidance on when to use bulk_search vs search_papers.
Search Google Scholar for multiple queries sequentially with rate limiting. Use this for batch literature reviews. Each query runs one at a time with delays between them to avoid being blocked. Args: queries: List of search queries (maximum 10). max_results_per_query: Results per query (1-10, default 5). year_from: Filter to papers published in or after this year. year_to: Filter to papers published in or before this year.
Get BibTeX citation entry for a paper. Args: title: Title of the paper to get BibTeX for.
Get papers that cite a given paper. Args: title: Title of the paper whose citations to retrieve. max_results: Maximum citing papers to return (1-20, default 10).
Get full details of a specific paper by its title. Args: title: The exact (or close) title of the paper.
Find a Google Scholar author profile by name. Args: name: Author name to search for (e.g. "Albert Einstein"). max_publications: Number of publications to include (1-20, default 5).
Output schemas not documented. Tools return Pydantic models (SearchResponse, AuthorResult, PaperDetails, etc.) but no inline schema documentation visible in source. LLMs cannot plan downstream calls without knowing response structure.
No tool annotations. All 6 tools are read-only (no state modification), but readOnlyHint is not declared. Agents cannot distinguish safe tools from destructive ones without explicit hints.
Error handling lacks recovery guidance. bulk_search catches exceptions and logs them, but error messages are generic ('Error: {e}'). No guidance on retryability, user-fixable vs fatal, or next steps.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 61 | <=2025-11-25 | v2 |
Search Google Scholar for academic papers by keyword. Args: query: Search query (same syntax as Google Scholar search box). max_results: Maximum number of results to return (1-20, default 10). year_from: Filter to papers published in or after this year. year_to: Filter to papers published in or before this year. sort_by: Sort order — "relevance" (default) or "date". include_patents: Include patents in results (default True).
Parameter constraints stated in descriptions but not enforced in schema. max_results documented as '1-20' but schema has no minimum/maximum. LLMs may pass invalid values; server clamps them silently.
Ambiguous tool selection guidance. bulk_search and search_papers both search papers; description says 'Use this for batch literature reviews' but does not explain when to prefer bulk_search over calling search_papers multiple times.