BioContextAI Knowledgebase MCP

Instructions

Implementation Reference

• src/biocontext_kb/core/clinicaltrials/_get_study_details.py:9-42 (handler)

  The handler function implementing bc_get_study_details (prefixed from get_study_details). Fetches detailed information for a clinical trial by NCT ID from ClinicalTrials.gov API v2, with input validation and error handling.
  @core_mcp.tool() def get_study_details( nct_id: Annotated[str, Field(description="NCT ID (e.g., 'NCT01234567')")], fields: Annotated[ str, Field(description="Comma-separated fields or 'all' for complete data. Default includes key modules.") ] = "IdentificationModule,StatusModule,SponsorCollaboratorsModule,DescriptionModule,ConditionsModule,DesignModule,ArmsInterventionsModule,OutcomesModule,EligibilityModule,ContactsLocationsModule", ) -> Union[Dict[str, Any], dict]: """Get complete trial details by NCT ID. Retrieves study design, eligibility, outcomes, locations, contacts, and metadata. Returns: dict: Study details with protocol sections including identification, status, sponsors, description, conditions, design, interventions, outcomes, eligibility, locations or error message. """ if not nct_id: return {"error": "NCT ID must be provided"} # Validate NCT ID format (should start with NCT followed by 8 digits) if not nct_id.upper().startswith("NCT") or len(nct_id) != 11: return {"error": "Invalid NCT ID format. Expected format: NCT12345678"} # Construct URL if fields.lower() == "all": url = f"https://clinicaltrials.gov/api/v2/studies/{nct_id}?format=json" else: url = f"https://clinicaltrials.gov/api/v2/studies/{nct_id}?fields={fields}&format=json" try: response = requests.get(url) response.raise_for_status() return response.json() except requests.exceptions.RequestException as e: if response.status_code == 404: return {"error": f"Study with NCT ID '{nct_id}' not found"} return {"error": f"Failed to fetch study details: {e!s}"}
• src/biocontext_kb/core/_server.py:1-6 (registration)

  Defines the core_mcp FastMCP server instance named 'BC'. Tool functions decorated with @core_mcp.tool() are registered here. Later imported into main app with 'bc' prefix.
  from fastmcp import FastMCP core_mcp = FastMCP( # type: ignore "BC", instructions="Provides access to biomedical knowledge bases.", )
• src/biocontext_kb/core/clinicaltrials/__init__.py:1-13 (registration)

  Imports the get_study_details function, triggering its @core_mcp.tool() decorator to register the tool in core_mcp.
  from ._get_recruiting_studies_by_location import get_recruiting_studies_by_location from ._get_studies_by_condition import get_studies_by_condition from ._get_studies_by_intervention import get_studies_by_intervention from ._get_study_details import get_study_details from ._search_studies import search_studies __all__ = [ "get_recruiting_studies_by_location", "get_studies_by_condition", "get_studies_by_intervention", "get_study_details", "search_studies", ]
• src/biocontext_kb/app.py:35-39 (registration)

  Imports core_mcp into the main mcp_app, prefixing all its tools with 'bc' (slugify('BC')). This creates the 'bc_get_study_details' tool in the final server.
  for mcp in [core_mcp, *(await get_openapi_mcps())]: await mcp_app.import_server( mcp, slugify(mcp.name), )
• src/biocontext_kb/core/__init__.py:7-7 (registration)

  Imports all clinicaltrials tools (including get_study_details) into core namespace, ensuring registration.
  from .clinicaltrials import *