LinkedIn MCP Server

Instructions

Implementation Reference

• linkedin_mcp_server/tools/person.py:37-115 (handler)

  The handler function that executes the tool logic: constructs LinkedIn profile URL from username, scrapes data using linkedin_scraper.Person, transforms nested objects (experiences, educations, etc.) into structured dictionaries, and returns comprehensive profile information with error handling.

  async def get_person_profile(linkedin_username: str) -> Dict[str, Any]:
      """
      Get a specific person's LinkedIn profile.
  
      Args:
          linkedin_username (str): LinkedIn username (e.g., "stickerdaniel", "anistji")
  
      Returns:
          Dict[str, Any]: Structured data from the person's profile
      """
      try:
          # Construct clean LinkedIn URL from username
          linkedin_url = f"https://www.linkedin.com/in/{linkedin_username}/"
  
          driver = safe_get_driver()
  
          logger.info(f"Scraping profile: {linkedin_url}")
          person = Person(linkedin_url, driver=driver, close_on_complete=False)
  
          # Convert experiences to structured dictionaries
          experiences: List[Dict[str, Any]] = [
              {
                  "position_title": exp.position_title,
                  "company": exp.institution_name,
                  "from_date": exp.from_date,
                  "to_date": exp.to_date,
                  "duration": exp.duration,
                  "location": exp.location,
                  "description": exp.description,
              }
              for exp in person.experiences
          ]
  
          # Convert educations to structured dictionaries
          educations: List[Dict[str, Any]] = [
              {
                  "institution": edu.institution_name,
                  "degree": edu.degree,
                  "from_date": edu.from_date,
                  "to_date": edu.to_date,
                  "description": edu.description,
              }
              for edu in person.educations
          ]
  
          # Convert interests to list of titles
          interests: List[str] = [interest.title for interest in person.interests]
  
          # Convert accomplishments to structured dictionaries
          accomplishments: List[Dict[str, str]] = [
              {"category": acc.category, "title": acc.title}
              for acc in person.accomplishments
          ]
  
          # Convert contacts to structured dictionaries
          contacts: List[Dict[str, str]] = [
              {
                  "name": contact.name,
                  "occupation": contact.occupation,
                  "url": contact.url,
              }
              for contact in person.contacts
          ]
  
          # Return the complete profile data
          return {
              "name": person.name,
              "about": person.about,
              "experiences": experiences,
              "educations": educations,
              "interests": interests,
              "accomplishments": accomplishments,
              "contacts": contacts,
              "company": person.company,
              "job_title": person.job_title,
              "open_to_work": getattr(person, "open_to_work", False),
          }
      except Exception as e:
          return handle_tool_error(e, "get_person_profile")

• linkedin_mcp_server/tools/person.py:29-36 (registration)

  The @mcp.tool decorator within register_person_tools registers the get_person_profile handler with metadata annotations defining the tool's title and hints.

  @mcp.tool(
      annotations=ToolAnnotations(
          title="Get Person Profile",
          readOnlyHint=True,
          destructiveHint=False,
          openWorldHint=True,
      )
  )

• linkedin_mcp_server/server.py:27-30 (registration)

  Top-level registration call to register_person_tools(mcp) in the create_mcp_server function, which defines and registers the get_person_profile tool.

  # Register all tools
  register_person_tools(mcp)
  register_company_tools(mcp)
  register_job_tools(mcp)

• linkedin_mcp_server/tools/person.py:37-37 (schema)

  Function signature defining input schema (linkedin_username: str) and output type (Dict[str, Any]).

  async def get_person_profile(linkedin_username: str) -> Dict[str, Any]:

• linkedin_mcp_server/cli.py:69-70 (registration)

  Lists 'get_person_profile' as a required tool in the generated Claude Desktop configuration.

  "requiredTools": [
      "get_person_profile",