ScholarScope MCP

Instructions

Implementation Reference

• src/server.py:181-243 (handler)

  The handler function decorated with @mcp.tool that implements the core logic for searching institutions via the OpenAlex API, including query sanitization, API request, result parsing into Institution models, pagination handling, and error management.

  @mcp.tool
  async def search_institutions(
          query: str,
          sort_by: Literal["relevance_score", "cited_by_count"] = "relevance_score",
          page: int = 1,
  ) -> PageResult:
      """
      Searches for institutions using the OpenAlex API.
  
      Args:
          query: The search name to look for the institutions.
          sort_by: The sorting criteria ("relevance_score" or "cited_by_count").
          page: The page number of the results to retrieve (default: 1).
  
      Returns:
          A JSON object containing a list of institutions+ids, or an error message if the search fails.
      """
      query = sanitize_search_text(query)
  
      params = {
          "filter": f"default.search:\"{query}\"",
          "sort": f"{sort_by}:desc",
          "page": page,
          "per_page": 10,
      }
  
      # 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 authors using: query={query}, sort_by={sort_by}, page={page}")
          try:
              result = await api.aget("/institutions", 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 institutions found with the query."
                  logger.info(error_message)
                  raise ToolError(error_message)
  
              # Successfully returns the searched papers
              institutions = Institution.from_list(result.get("results", []) or [])
              success_message = f"Found {len(institutions)} institution(s)."
              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=Institution.list_to_json(institutions),
                  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:5-45 (schema)

  Pydantic BaseModel for Institution, including parsing from OpenAlex JSON responses and serialization methods used in the search_institutions output.

  class Institution(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
      )
  
      name: str
      id: Optional[str] = None
  
      @classmethod
      def from_json(cls, json_obj: Dict[str, Any]) -> "Institution":
          inst_name = ""
          inst_id = None
  
          if "institution" in json_obj:
              institution = json_obj.get("institution", {}) or {}
              inst_name = institution.get("display_name", "") or ""
              inst_id = institution.get("id")
          elif "raw_affiliation_string" in json_obj:
              inst_name = json_obj.get("raw_affiliation_string", "") or ""
              ids = json_obj.get("institution_ids")
              if ids and len(ids) >= 1:
                  inst_id = ids[0]
          elif "id" in json_obj:
              inst_name = json_obj.get("display_name", "")
              inst_id = json_obj.get("id")
  
          return cls(name=inst_name, id=inst_id)
  
      @classmethod
      def from_list(cls, json_list: List[dict]) -> List["Institution"]:
          return [cls.from_json(item) for item in json_list]
  
      @staticmethod
      def list_to_json(institutions: List["Institution"]) -> List[dict]:
          return [institution.model_dump(exclude_none=True) for institution in institutions]
  
      def __str__(self) -> str:
          return self.model_dump_json(exclude_none=True)

• src/schemas.py:133-139 (schema)

  Pydantic model defining the structure of paginated results returned by the search_institutions tool.

  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/utils.py:4-10 (helper)

  Utility function to sanitize search queries by removing commas and normalizing whitespace, called at the start of search_institutions.

  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