read_section
Extract specific sections from documents by providing document and section identifiers to retrieve targeted content with subsections for focused analysis.
Instructions
Read content of a specific document section.
Args:
doc_id: Document identifier
section_id: Section ID from outline (e.g., 'h1_0', 'h2_1')
Returns:
Section content with subsections
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| doc_id | Yes | ||
| section_id | Yes |
Implementation Reference
- server.py:70-82 (handler)MCP tool handler for 'read_section'. Decorated with @mcp.tool(), handles input parameters, strips '#' from section_id if present, and delegates to Navigator.read_section. This is the entry point for the tool.@mcp.tool() def read_section(doc_id: str, section_id: str) -> str: """Read content of a specific document section. Args: doc_id: Document identifier section_id: Section ID from outline (e.g., 'h1_0', 'h2_1') Returns: Section content with subsections """ return navigator.read_section(doc_id, section_id.strip("#"))
- docnav/navigator.py:852-872 (helper)Core helper method in Navigator class that retrieves the document, finds the specified section node, collects content from the node and all its descendants recursively, and joins them into a string. Handles error cases for missing document or section.def read_section(self, doc_id: str, section_id: str) -> str: """Read specified section content.""" document = self.get_document(doc_id) if not document: return f"Document '{doc_id}' not found" # Get the node and return its content with subsections node = document.get_node(section_id) if not node: return f"Section '{section_id}' not found" content = [node.content] if node.content else [] def collect_content(n: DocumentNode) -> None: for child in n.children: if child.content: content.append(child.content) collect_content(child) collect_content(node) return "\n".join(content)
- server.py:70-70 (registration)The @mcp.tool() decorator registers the read_section function as an MCP tool.@mcp.tool()