ScholarScope MCP

Instructions

Implementation Reference

• src/server.py:36-110 (handler)

  The core handler function for the 'search_papers' tool, decorated with @mcp.tool for registration. It constructs API parameters, queries OpenAlex API for papers, processes results using Work models, and returns a PageResult.
  @mcp.tool async def search_papers( query: str, search_by: Literal["default", "title", "title_and_abstract"] = "default", sort_by: Literal["relevance_score", "cited_by_count", "publication_date"] = "relevance_score", institution_name: Optional[str] = None, author_id: Optional[str] = None, page: int = 1, ) -> PageResult: """ Searches for academic papers using the OpenAlex API. Args: query: The search term or keywords to look for in the papers. search_by: The field to search in ("default", "title", or "title_and_abstract"). sort_by: The sorting criteria ("relevance_score", "cited_by_count", or "publication_date"). institution_name: An optional institution or affiliation name to filter search results. author_id: An optional OpenAlex Author ID to filter search results. e.g., "https://openalex.org/A123456789" page: The page number of the results to retrieve (default: 1). Returns: A JSON object containing a list of searched papers+ids, or an error message if the search fails. """ query = sanitize_search_text(query) institution_name = sanitize_search_text(institution_name) params = { "filter": f"{search_by}.search:\"{query}\"", "sort": f"{sort_by}:desc", "page": page, "per_page": 10, } if institution_name: params["filter"] += f",raw_affiliation_strings.search:\"{institution_name}\"" if author_id: params["filter"] += f",authorships.author.id:{author_id}" # Fetches search results from the OpenAlex API async with RequestAPI("https://api.openalex.org", default_params={"mailto": OPENALEX_MAILTO}) as api: logger.info(f"Searching for papers using: query={query}, search_by={search_by}, sort_by={sort_by}, page={page}") try: result = await api.aget("/works", params=params) # Returns a message for when the search results are empty if result is None or len(result.get("results", []) or []) == 0: error_message = "No works found with the query." logger.info(error_message) raise ToolError(error_message) # Successfully returns the searched papers works = Work.from_list(result.get("results", []) or []) success_message = f"Found {len(works)} papers." logger.info(success_message) total_count = (result.get("meta", {}) or {}).get("count") if total_count and total_count > params["per_page"] * params["page"]: has_next = True else: has_next = None return PageResult( data=Work.list_to_json(works), total_count=total_count, per_page=params["per_page"], page=params["page"], has_next=has_next ) except httpx.HTTPStatusError as e: error_message = f"Request failed with status: {e.response.status_code}" logger.error(error_message) raise ToolError(error_message) except httpx.RequestError as e: error_message = f"Network error: {str(e)}" logger.error(error_message) raise ToolError(error_message)
• src/schemas.py:133-139 (schema)

  Pydantic model defining the output schema for search_papers (and similar paginated tools), including data list, pagination metadata.
  class PageResult(BaseModel): data: List[Union[Institution, Author, Work, dict]] = Field(default_factory=list) total_count: Optional[int] = None per_page: int page: int has_next: Optional[bool] = None
• src/schemas.py:82-131 (schema)

  Pydantic model for Work (paper) objects, used in search_papers results data. Includes parsing from OpenAlex JSON and serialization.
  class Work(BaseModel): model_config = ConfigDict( frozen=False, # set True for immutability validate_assignment=True, # runtime type safety on attribute set str_strip_whitespace=True, # trims incoming strings ) title: str = None ids: Dict[str, str] = Field(default_factory=dict) cited_by_count: Optional[int] = None authors: List[Author] = Field(default_factory=list) publication_date: Optional[str] = None preferred_fulltext_url: Optional[str] = None @classmethod def from_json(cls, json_obj: Dict[str, Any]) -> "Work": # Gets title and page urls title = json_obj.get("title") or json_obj.get("display_name") or "" # Prioritize Open Access url preferred_fulltext_url = (json_obj.get("best_oa_location", {}) or {}).get("pdf_url") if preferred_fulltext_url is None: preferred_fulltext_url = (json_obj.get("best_oa_location", {}) or {}).get("landing_page_url") if preferred_fulltext_url is None: preferred_fulltext_url = (json_obj.get("primary_location", {}) or {}).get("pdf_url") if preferred_fulltext_url is None: preferred_fulltext_url = (json_obj.get("primary_location", {}) or {}).get("landing_page_url") # Gets individual authors of the work authors = Author.from_list(json_obj.get("authorships", []) or []) return cls( title=title, ids=json_obj.get("ids", {}) or {}, authors=authors, cited_by_count=json_obj.get("cited_by_count"), publication_date=json_obj.get("publication_date"), preferred_fulltext_url=preferred_fulltext_url ) @classmethod def from_list(cls, json_list: List[dict]) -> List["Work"]: return [cls.from_json(item) for item in json_list] @staticmethod def list_to_json(works: List["Work"]) -> List[dict]: return [work.model_dump(exclude_none=True) for work in works] def __str__(self) -> str: return self.model_dump_json(exclude_none=True)
• src/utils.py:4-10 (helper)

  Utility function to sanitize search query text by removing commas and normalizing whitespace, used in search_papers.
  def sanitize_search_text(s: str) -> str: """Remove commas and collapse whitespace for API search terms for OpenAlex to work""" if not s: return s s = s.replace(",", " ") s = re.sub(r"\s+", " ", s).strip() return s