Freshdesk MCP server

Instructions

Implementation Reference

• src/freshdesk_mcp/server.py:176-223 (handler)

  The core handler function for the 'get_tickets' MCP tool, decorated with @mcp.tool() for registration. It fetches tickets from the Freshdesk API endpoint /api/v2/tickets, supports pagination via page and per_page parameters, validates inputs, parses the Link header for pagination info using the helper parse_link_header, and returns tickets with pagination metadata or error messages.
  @mcp.tool() async def get_tickets(page: Optional[int] = 1, per_page: Optional[int] = 30) -> Dict[str, Any]: """Get tickets from Freshdesk with pagination support.""" # Validate input parameters if page < 1: return {"error": "Page number must be greater than 0"} if per_page < 1 or per_page > 100: return {"error": "Page size must be between 1 and 100"} url = f"https://{FRESHDESK_DOMAIN}/api/v2/tickets" params = { "page": page, "per_page": per_page } headers = { "Authorization": f"Basic {base64.b64encode(f'{FRESHDESK_API_KEY}:X'.encode()).decode()}", "Content-Type": "application/json" } async with httpx.AsyncClient() as client: try: response = await client.get(url, headers=headers, params=params) response.raise_for_status() # Parse pagination from Link header link_header = response.headers.get('Link', '') pagination_info = parse_link_header(link_header) tickets = response.json() return { "tickets": tickets, "pagination": { "current_page": page, "next_page": pagination_info.get("next"), "prev_page": pagination_info.get("prev"), "per_page": per_page } } except httpx.HTTPStatusError as e: return {"error": f"Failed to fetch tickets: {str(e)}"} except Exception as e: return {"error": f"An unexpected error occurred: {str(e)}"}
• src/freshdesk_mcp/server.py:21-52 (helper)

  Supporting utility function that parses the Freshdesk API's Link header to extract 'next' and 'prev' pagination page numbers. Called within the get_tickets handler to provide pagination metadata in the response.
  def parse_link_header(link_header: str) -> Dict[str, Optional[int]]: """Parse the Link header to extract pagination information. Args: link_header: The Link header string from the response Returns: Dictionary containing next and prev page numbers """ pagination = { "next": None, "prev": None } if not link_header: return pagination # Split multiple links if present links = link_header.split(',') for link in links: # Extract URL and rel match = re.search(r'<(.+?)>;\s*rel="(.+?)"', link) if match: url, rel = match.groups() # Extract page number from URL page_match = re.search(r'page=(\d+)', url) if page_match: page_num = int(page_match.group(1)) pagination[rel] = page_num return pagination
• src/freshdesk_mcp/server.py:176-176 (registration)

  The @mcp.tool() decorator registers the get_tickets function as an MCP tool.
  @mcp.tool()