Weibo MCP Server

Instructions

Get comments for a specific Weibo post.
    
Returns:
    list[dict]: List of dictionaries containing comments

Implementation Reference

• src/mcp_server_weibo/server.py:143-155 (handler)

  MCP tool handler and registration for 'get_comments'. Defines input schema via Annotated Fields and delegates execution to WeiboCrawler.get_comments.

  @mcp.tool()
  async def get_comments(
      ctx: Context, 
      feed_id: Annotated[int, Field(description="The unique identifier of the Weibo post")], 
      page: Annotated[int, Field(description="Page number for pagination, defaults to 1", default=1)] = 1
      ) -> list[dict]:
      """
      Get comments for a specific Weibo post.
          
      Returns:
          list[dict]: List of dictionaries containing comments
      """
      return await crawler.get_comments(feed_id, page)

• src/mcp_server_weibo/weibo.py:258-278 (handler)

  Core implementation of the get_comments functionality in WeiboCrawler. Performs HTTP GET to Weibo comments API, parses JSON response, and converts comments using helper.

  async def get_comments(self, feed_id: str, page: int = 1) -> list[CommentItem]:
      """
      Get comments for a specific Weibo post.
  
      Args:
          feed_id (str): The ID of the Weibo post
          page (int): The page number for pagination, defaults to 1
  
      Returns:
          list[CommentItem]: List of comments for the specified Weibo post
      """
      try:
          async with httpx.AsyncClient() as client:
              url = COMMENTS_URL.format(feed_id=feed_id, page=page)
              response = await client.get(url, headers=DEFAULT_HEADERS)
              data = response.json()
              comments = data.get('data', {}).get('data', [])
              return [self._to_comment_item(comment) for comment in comments]
      except httpx.HTTPError:
          self.logger.error(f"Unable to fetch comments for feed_id '{feed_id}'", exc_info=True)
          return []

• src/mcp_server_weibo/schemas.py:90-108 (schema)

  Pydantic BaseModel defining the structure of individual comment items returned by the tool.

  class CommentItem(BaseModel):
      """
      Data model for a single comment on a Weibo post.
      
      Attributes:
          id (int): Unique identifier for the comment
          text (str): Content of the comment
          created_at (str): Timestamp when the comment was created
          user (UserProfile): User information associated with the comment
          like_count (int): Number of likes on the comment
          reply_count (int): Number of replies to the comment
      """
      id: int = Field()
      text: str = Field()
      created_at: str = Field()
      source: str = Field()
      user: UserProfile = Field()
      reply_id: Union[int, None] = Field(default=None)
      reply_text: str = Field(default="")

• src/mcp_server_weibo/weibo.py:495-513 (helper)

  Supporting utility function that transforms raw dictionary from Weibo API into a structured CommentItem model.

  def _to_comment_item(self, item: dict) -> CommentItem:
      """
      Convert raw comment data to CommentItem object.
  
      Args:
          item (dict): Raw comment data from Weibo API
  
      Returns:
          CommentItem: Formatted comment information
      """
      return CommentItem(
          id = item.get('id'),
          text = item.get('text'),
          created_at = item.get('created_at'),
          user = self._to_user_profile(item.get('user', {})),
          source=item.get('source', ''),
          reply_id = item.get('reply_id', None),
          reply_text = item.get('reply_text', ''),
      )

• src/mcp_server_weibo/consts.py:17-20 (helper)

  Constant URL template used to construct the API endpoint for retrieving post comments.

  # URL template for fetching comments of a specific Weibo post
  # {feed_id}: The ID of the Weibo post
  # {page}: The page number for pagination
  COMMENTS_URL = 'https://m.weibo.cn/api/comments/show?id={feed_id}&page={page}'