WSB Analyst MCP Server

Instructions

Implementation Reference

• server.py:464-517 (handler)

  The handler function decorated with @mcp.tool(), which implements the get_external_links tool. It fetches top WSB posts using find_top_posts, retrieves detailed post data including extracted external links using fetch_batch_post_details, collects and deduplicates all unique external links from posts and comments, and returns them sorted.
  @mcp.tool() async def get_external_links(min_score: int = 100, min_comments: int = 10, limit: int = 10, ctx: Context = None) -> dict: """ Get all external links from top WSB posts. Args: min_score: Minimum score (upvotes) required min_comments: Minimum number of comments required limit: Maximum number of posts to scan Returns: Dictionary with all unique external links found """ if ctx: await ctx.report_progress(0, 3) # Get filtered posts posts_result = await find_top_posts(min_score, min_comments, limit) if "error" in posts_result: return {"error": posts_result["error"]} if len(posts_result["posts"]) == 0: return {"count": 0, "links": []} # Collect post IDs post_ids = [post["id"] for post in posts_result["posts"]] if ctx: await ctx.report_progress(1, 3) # Get details for all posts details_result = await fetch_batch_post_details(post_ids) if "error" in details_result: return {"error": details_result["error"]} # Extract all links all_links = [] for post_id, post_detail in details_result["posts"].items(): if "extracted_links" in post_detail: all_links.extend(post_detail["extracted_links"]) if ctx: await ctx.report_progress(2, 3) # Remove duplicates and sort unique_links = sorted(list(set(all_links))) if ctx: await ctx.report_progress(3, 3) return { "count": len(unique_links), "links": unique_links }
• server.py:69-80 (helper)

  Helper function that extracts potential URL links from text using regex, excluding common Reddit domains, and filters them further using is_valid_external_link.
  def extract_valid_links(text: str) -> list[str]: if not text: return [] url_pattern = re.compile( r'https?://(?!(?:www\.)?reddit\.com|(?:www\.)?i\.redd\.it|(?:www\.)?v\.redd\.it|(?:www\.)?imgur\.com|' r'(?:www\.)?preview\.redd\.it|(?:www\.)?sh\.reddit\.com|[^.]*\.reddit\.com)' r'[^\s)\]}"\']+', re.IGNORECASE ) links = url_pattern.findall(text) return [link for link in links if is_valid_external_link(link)]
• server.py:58-67 (helper)

  Helper function that determines if a URL is a valid external link by checking against a list of excluded domains like Reddit, Imgur, etc.
  def is_valid_external_link(url: str) -> bool: excluded_domains = [ "reddit.com", "redd.it", "imgur.com", "gfycat.com", "redgifs.com", "giphy.com", "imgflip.com", "youtu.be", "discord.gg", ] if any(domain in url for domain in excluded_domains): return False return True
• server.py:464-464 (registration)

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