Docs-MCP

Instructions

Args:
    pattern: 検索パターン（正規表現対応）
    ignore_case: 大文字小文字を無視するか（デフォルト: True）

Implementation Reference

• src/mcp_server_docs/server.py:43-51 (handler)

  MCP tool handler for 'grep_docs'. Registers the tool via @mcp.tool() decorator and delegates execution to DocumentManager.grep_search method.

  @mcp.tool()
  async def grep_docs(pattern: str, ignore_case: bool = True) -> str:
      """ドキュメント内をgrepで検索
  
      Args:
          pattern: 検索パターン（正規表現対応）
          ignore_case: 大文字小文字を無視するか（デフォルト: True）
      """
      return doc_manager.grep_search(pattern, ignore_case)

• src/mcp_server_docs/document_manager.py:269-296 (helper)

  Core implementation of the grep search functionality. Compiles the regex pattern, searches all loaded documents line-by-line, collects matches with file path and line number, limits to 100 results, and formats the output.

  def grep_search(self, pattern: str, ignore_case: bool = True) -> str:
      """正規表現でドキュメントを検索"""
      try:
          flags = re.IGNORECASE if ignore_case else 0
          regex = re.compile(pattern, flags)
      except re.error as e:
          return f"Error: Invalid regex pattern: {e}"
  
      results = []
      for doc_path, content in sorted(self.docs_content.items()):
          lines = content.split("\n")
          for i, line in enumerate(lines, 1):
              if regex.search(line):
                  line_preview = line.strip()
                  if len(line_preview) > 120:
                      line_preview = line_preview[:117] + "..."
                  results.append(f"{doc_path}:{i}: {line_preview}")
  
      if not results:
          return "No matches found"
  
      # 結果が多すぎる場合は制限
      if len(results) > 100:
          total = len(results)
          results = results[:100]
          results.append(f"\n... and {total - 100} more matches")
  
      return "\n".join(results)