RBT Document Editor

Get document outline (structure without blocks). **BEST PRACTICE: Always use this FIRST before editing documents!** - See what sections exist - Find section/block IDs for editing - Saves tokens by not loading full content Returns lightweight document structure with metadata, info, and section tree. Significantly reduces token consumption compared to reading full document. Args: project_id: Project identifier feature_id: Feature identifier (for RBT documents) doc_type: Document type - REQ/BP/TASK (for RBT documents) file_path: File path relative to docs/ (for general documents) **For TASK documents: supports fuzzy matching! Just provide the index number.** Returns: Document outline JSON with metadata, info, title, and sections Example: # RBT document get_outline(project_id="knowledge-smith", feature_id="rbt-mcp-tool", doc_type="BP") # TASK document (FUZZY SEARCH - just use the index!) get_outline(project_id="knowledge-smith", feature_id="rbt-mcp-tool", doc_type="TASK", file_path="001") # Matches TASK-001-*.md # General document get_outline(project_id="knowledge-smith", file_path="architecture/overview.md") @REQ: REQ-rbt-mcp-tool @BP: BP-rbt-mcp-tool @TASK: TASK-005-GetOutlineTool-Server-Setup

Read section or block by ID. **KEY FEATURE: Read ONLY what you need!** - Read a single block (blk-*) → saves tokens, don't read the whole section - Read a section (sec-*) → gets section + all its blocks **TIP: Use get_outline() first to find available IDs** Automatically determines content type based on ID prefix: - sec-* → reads section with blocks and nested sections - blk-* → reads single block data Args: project_id: Project identifier content_id: Content ID (section or block ID starting with 'sec-' or 'blk-') feature_id: Feature identifier (for RBT documents) doc_type: Document type - REQ/BP/TASK (for RBT documents) file_path: File path relative to docs/ (for general documents) Returns: Content data (section or block) Example: # Read section read_content( project_id="knowledge-smith", feature_id="rbt-mcp-tool", doc_type="BP", content_id="sec-components" ) # Read block read_content( project_id="knowledge-smith", feature_id="rbt-mcp-tool", doc_type="BP", content_id="blk-component-table" ) @REQ: REQ-rbt-mcp-tool @BP: BP-rbt-mcp-tool @TASK: TASK-006-ReadSectionTool

Update section summary text. Args: project_id: Project identifier section_id: Section ID to update new_summary: New summary text feature_id: Feature identifier (for RBT documents) doc_type: Document type - REQ/BP/TASK (for RBT documents) file_path: File path relative to docs/ (for general documents) Returns: Success message Example: update_section_summary( project_id="knowledge-smith", feature_id="rbt-mcp-tool", doc_type="BP", section_id="sec-components", new_summary="Updated component specifications with new requirements" ) @REQ: REQ-rbt-mcp-tool @BP: BP-rbt-mcp-tool @TASK: TASK-007-UpdateSectionSummaryTool-Server-Setup

Create new section under specified parent. Note: For RBT documents, cannot create root sections (only sub-sections). Args: project_id: Project identifier parent_id: Parent section ID title: Section title summary: Optional section summary feature_id: Feature identifier (for RBT documents) doc_type: Document type - REQ/BP/TASK (for RBT documents) file_path: File path relative to docs/ (for general documents) Returns: New section ID Example: create_section( project_id="knowledge-smith", feature_id="rbt-mcp-tool", doc_type="REQ", parent_id="sec-use-cases", title="Additional Use Cases", summary="Extended scenarios for document editing" ) @REQ: REQ-rbt-mcp-tool @BP: BP-rbt-mcp-tool @TASK: TASK-008-CreateSectionTool

Create new block in specified section. **TIP: Block IDs are auto-generated, but you can identify blocks later by their content. Consider adding meaningful content/titles to make blocks easier to find with get_outline().** Args: project_id: Project identifier section_id: Section ID to add block to block_type: Block type - paragraph/code/list/table content: Content for paragraph/code blocks items: Items for list blocks language: Language for code blocks header: Header row for table blocks rows: Data rows for table blocks feature_id: Feature identifier (for RBT documents) doc_type: Document type - REQ/BP/TASK (for RBT documents) file_path: File path relative to docs/ (for general documents) Returns: New block ID Example: # Create paragraph create_block( project_id="knowledge-smith", file_path="docs/guide.md", section_id="sec-intro", block_type="paragraph", content="This is a new paragraph." ) # Create list create_block( project_id="knowledge-smith", file_path="docs/guide.md", section_id="sec-features", block_type="list", items=["Feature 1", "Feature 2"] ) @REQ: REQ-rbt-mcp-tool @BP: BP-rbt-mcp-tool @TASK: TASK-009-CreateBlockTool

Update existing block content based on block type. Supports updating different block types: - paragraph: content - code: content, language (optional) - list: title (optional), items - table: header, rows Args: project_id: Project identifier block_id: Block ID to update content: New content for paragraph/code blocks title: New title for list blocks items: New items for list blocks language: Programming language for code blocks header: New header for table blocks rows: New rows for table blocks feature_id: Feature identifier (for RBT documents) doc_type: Document type - REQ/BP/TASK (for RBT documents) file_path: File path relative to docs/ (for general documents) Returns: Success message dictionary Example: # Update paragraph block update_block( project_id="knowledge-smith", file_path="docs/guide.md", block_id="blk-paragraph-1", content="Updated paragraph content." ) # Update list block update_block( project_id="knowledge-smith", feature_id="rbt-mcp-tool", doc_type="REQ", block_id="blk-requirements", title="**Updated Requirements**", items=["Req 1", "Req 2", "Req 3"] ) # Update code block update_block( project_id="knowledge-smith", file_path="docs/examples.md", block_id="blk-code-sample", content="def hello():\n print('Hello')", language="python" ) # Update table block update_block( project_id="knowledge-smith", feature_id="test", doc_type="BP", block_id="blk-table-data", header=["Name", "Value"], rows=[["Item1", "100"], ["Item2", "200"]] ) @REQ: REQ-rbt-mcp-tool @BP: BP-rbt-mcp-tool @TASK: TASK-010-UpdateBlockTool

Delete block from document. Args: project_id: Project identifier block_id: Block ID to delete feature_id: Feature identifier (for RBT documents) doc_type: Document type - REQ/BP/TASK (for RBT documents) file_path: File path relative to docs/ (for general documents) Returns: Success message Example: delete_block( project_id="knowledge-smith", file_path="docs/guide.md", block_id="blk-paragraph-2" ) @REQ: REQ-rbt-mcp-tool @BP: BP-rbt-mcp-tool @TASK: TASK-011-DeleteBlockTool

Append item to list block. Args: project_id: Project identifier block_id: List block ID item: Item text to append feature_id: Feature identifier (for RBT documents) doc_type: Document type - REQ/BP/TASK (for RBT documents) file_path: File path relative to docs/ (for general documents) Returns: Success message Example: append_list_item( project_id="knowledge-smith", feature_id="rbt-mcp-tool", doc_type="REQ", block_id="blk-func-req-list", item="New functional requirement item" ) @REQ: REQ-rbt-mcp-tool @BP: BP-rbt-mcp-tool @TASK: TASK-012-AppendListItemTool

Update specific table row. Args: project_id: Project identifier block_id: Table block ID row_index: Row index to update (0-based, excluding header) row_data: New row data (list of cell values) feature_id: Feature identifier (for RBT documents) doc_type: Document type - REQ/BP/TASK (for RBT documents) file_path: File path relative to docs/ (for general documents) Returns: Success message Example: update_table_row( project_id="knowledge-smith", feature_id="rbt-mcp-tool", doc_type="BP", block_id="blk-component-spec-table", row_index=0, row_data=["PathResolver", "Updated description", "new input", "new output"] ) @REQ: REQ-rbt-mcp-tool @BP: BP-rbt-mcp-tool @TASK: TASK-013-UpdateTableRowTool

Append a new row to a table block. Args: project_id: Project identifier block_id: Table block ID row_data: Row data (list of cell values) feature_id: Feature identifier (for RBT documents) doc_type: Document type - REQ/BP/TASK (for RBT documents) file_path: File path relative to docs/ (for general documents) Returns: Success message Example: append_table_row( project_id="knowledge-smith", feature_id="rbt-mcp-tool", doc_type="BP", block_id="blk-component-spec-table", row_data=["new_tool", "New tool description", "inputs", "outputs", "TASK-XXX", "Acceptance criteria"] ) @REQ: REQ-rbt-mcp-tool @BP: BP-rbt-mcp-tool @TASK: TASK-017-AppendTableRowTool

Create new document from template with placeholder replacement. Creates a new document by loading the appropriate template (Task/Blueprint/Requirement), auto-filling common placeholders, and saving as .new.md with complete structure. Auto-filled placeholders (from parameters): - [project-id]: From project_id parameter - [feature-id]: From feature_id parameter - [feature-name]: From feature_id parameter - [YYYY-MM-DD]: Current date (auto-generated) Custom placeholders (need to be in replacements): - [task-name]: Task name (for Task type) - [任務標題]/[需求標題]/[藍圖標題]: Document titles - Any other template-specific placeholders Args: project_id: Project identifier (auto-fills [project-id]) doc_type: Document type - "Task", "Blueprint", "Requirement" for RBT documents, or custom types like "General", "Architecture", "Guide", "API" for general documents. doc_type serves as classification for better document organization and retrieval. replacements: Dictionary of custom placeholder -> value mappings feature_id: Feature identifier (auto-fills [feature-id] and [feature-name]) file_path: File path within docs/ (for general documents) Returns: Success message with created file path Example (Task document): create_document( project_id="knowledge-smith", doc_type="Task", feature_id="rbt-mcp-tool", replacements={ "task-name": "PathResolver", "任務標題": "實作 PathResolver 路徑解析與驗證" } ) # project-id, feature-id, feature-name, date 會自動填入 Example (Blueprint document): create_document( project_id="knowledge-smith", doc_type="Blueprint", feature_id="rbt-mcp-tool", replacements={ "藍圖標題": "RBT 文件編輯 MCP Tool" } ) @REQ: REQ-rbt-mcp-tool @BP: BP-rbt-mcp-tool @TASK: TASK-014-CreateDocumentTool

Update info section fields (status, update_date, dependencies). Supports partial updates - only provided fields are updated, others remain unchanged. At least one field must be provided for update. Args: project_id: Project identifier status: Optional new status value update_date: Optional new update_date value (format: YYYY-MM-DD) dependencies: Optional new dependencies list feature_id: Feature identifier (for RBT documents) doc_type: Document type - REQ/BP/TASK (for RBT documents) file_path: File path relative to docs/ (for general documents) Returns: Success message Example: # Update status only update_info( project_id="knowledge-smith", feature_id="rbt-mcp-tool", doc_type="TASK", file_path="017", status="Done" ) # Update multiple fields update_info( project_id="knowledge-smith", feature_id="rbt-mcp-tool", doc_type="BP", status="In Progress", update_date="2025-10-08", dependencies=["TASK-001", "TASK-002"] ) @REQ: REQ-rbt-mcp-tool @BP: BP-rbt-mcp-tool @TASK: TASK-016-UpdateInfoTool

Clear document cache. Args: file_path: Optional file path to clear; if None, clears all cache Returns: Success message Example: # Clear specific file clear_cache(file_path="/path/to/document.md") # Clear all cache clear_cache() @REQ: REQ-rbt-mcp-tool @BP: BP-rbt-mcp-tool @TASK: TASK-015-ClearCacheTool-Server-Setup