Oblique Strategies MCP Server

Instructions

Implementation Reference

• oblique_strategies_mcp/strategies_core.py:82-105 (handler)

  The actual business logic for the search_strategies tool, which filters strategies by query and edition.

  def search_strategies(
      self, query: str, edition: Optional[str] = None
  ) -> Dict[str, Any]:
      """Search for strategies containing the specified keywords."""
      query_lower = query.lower()
      results = []
  
      # Determine which editions to search
      editions_to_search = (
          [edition] if edition and edition in self.EDITIONS else self.EDITIONS.keys()
      )
  
      for ed in editions_to_search:
          try:
              strategies = self.load_strategies(ed)
              for strategy in strategies:
                  if query_lower in strategy.lower():
                      results.append({"strategy": strategy, "edition": ed})
          except (FileNotFoundError, Exception):
              # Skip editions that can't be loaded
              continue
  
      return {
          "query": query,

• oblique_strategies_mcp/__main__.py:35-48 (registration)

  MCP tool registration and entry point for search_strategies.

  @mcp.tool()
  def search_strategies(query: str, edition: Optional[str] = None) -> Dict[str, Any]:
      """
      Search for strategies containing the specified keywords.
  
      Args:
          query: Keywords to search for (case-insensitive).
          edition: Optional edition to limit search to. If not specified,
                  searches all editions.
  
      Returns:
          A dictionary containing matching strategies and their editions.
      """
      return manager.search_strategies(query, edition)