BioContextAI Knowledgebase MCP

Instructions

Implementation Reference

• src/biocontext_kb/core/grants/_search_grants_gov.py:9-55 (handler)

  The primary handler function for the 'search_grants_gov' tool (note: tool name is 'search_grants_gov', not 'bc_search_grants_gov'). Decorated with @core_mcp.tool() for MCP registration. Includes input schema via Pydantic Annotated Fields and full implementation of the grants.gov API search logic.

  @core_mcp.tool()
  def search_grants_gov(
      keyword: Annotated[Optional[str], Field(description="Search keyword")] = None,
      opp_num: Annotated[Optional[str], Field(description="Opportunity number")] = None,
      eligibilities: Annotated[Optional[str], Field(description="Eligibilities (comma-separated)")] = None,
      agencies: Annotated[Optional[str], Field(description="Agency codes (comma-separated)")] = None,
      rows: Annotated[int, Field(description="Results to return")] = 10,
      opp_statuses: Annotated[
          Optional[str], Field(description="'forecasted|posted' (pipe-separated, default: 'forecasted|posted')")
      ] = "forecasted|posted",
      aln: Annotated[Optional[str], Field(description="Assistance Listing Number")] = None,
      funding_categories: Annotated[Optional[str], Field(description="Categories (comma-separated)")] = None,
  ) -> dict:
      """Search grants.gov by keyword, agency, or other criteria. Returns opportunity listings with deadlines and eligibility.
  
      Returns:
          dict: Grant opportunities list with titles, agencies, deadlines, funding amounts, eligibility criteria or error message.
      """
      url = "https://api.grants.gov/v1/api/search2"
  
      # Build request payload
      payload = {"rows": rows, "oppStatuses": opp_statuses or "forecasted|posted"}
  
      # Add optional parameters if provided
      if keyword:
          payload["keyword"] = keyword
      if opp_num:
          payload["oppNum"] = opp_num
      if eligibilities:
          payload["eligibilities"] = eligibilities
      if agencies:
          payload["agencies"] = agencies
      if aln:
          payload["aln"] = aln
      if funding_categories:
          payload["fundingCategories"] = funding_categories
  
      try:
          headers = {"Content-Type": "application/json"}
          response = requests.post(url, json=payload, headers=headers)
          response.raise_for_status()
  
          # Return the JSON response
          return response.json()
      except requests.exceptions.RequestException as e:
          return {"error": f"Failed to fetch grants data: {e!s}"}

• src/biocontext_kb/core/__init__.py:10-10 (registration)

  Wildcard import statement that brings the search_grants_gov tool into the core module. Since the function is decorated with @core_mcp.tool(), this import registers the tool with the core_mcp FastMCP instance.

  from .grants._search_grants_gov import *

• src/biocontext_kb/core/grants/__init__.py:1-1 (registration)

  Re-export of the search_grants_gov function from its implementation module, facilitating the wildcard import in core/__init__.py.

  from ._search_grants_gov import search_grants_gov

• src/biocontext_kb/core/grants/_search_grants_gov.py:10-21 (schema)

  Input schema defined via Pydantic Field descriptions in the function signature parameters.

  def search_grants_gov(
      keyword: Annotated[Optional[str], Field(description="Search keyword")] = None,
      opp_num: Annotated[Optional[str], Field(description="Opportunity number")] = None,
      eligibilities: Annotated[Optional[str], Field(description="Eligibilities (comma-separated)")] = None,
      agencies: Annotated[Optional[str], Field(description="Agency codes (comma-separated)")] = None,
      rows: Annotated[int, Field(description="Results to return")] = 10,
      opp_statuses: Annotated[
          Optional[str], Field(description="'forecasted|posted' (pipe-separated, default: 'forecasted|posted')")
      ] = "forecasted|posted",
      aln: Annotated[Optional[str], Field(description="Assistance Listing Number")] = None,
      funding_categories: Annotated[Optional[str], Field(description="Categories (comma-separated)")] = None,
  ) -> dict: