Fixes type mismatch error preventing MCP clients from calling modify_sheet_values.
## Problem
FastMCP 2.11.1 passes complex nested types (like List[List[str]]) as JSON strings
through the MCP protocol, but the schema validation expected the actual Python type.
This caused validation failures with error: '[["Task"]]' is not valid under any of
the given schemas.
## Solution
- Accept both str and List[List[str]] types for the values parameter
- Parse JSON string input to Python list when received from MCP clients
- Validate structure is a proper 2D array (list of lists)
- Maintain backward compatibility with direct Python list input
## Changes
- Added json import and Union type hint
- Modified values parameter to accept Union[str, List[List[str]]]
- Added JSON parsing logic with proper validation
- Updated docstring to document both input formats
This follows the proven pattern used in calendar_tools.py for handling reminders.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Add full document editing and manipulation support to Google Docs tools,
expanding beyond read-only operations to match Google Docs API capabilities.
Features added:
- Text operations: update_doc_text, find_and_replace_doc, format_doc_text
- Structural elements: insert_doc_elements, insert_doc_image
- Document management: update_doc_headers_footers, batch_update_doc
- Helper functions module for common document operations
Changes:
- gdocs/docs_tools.py: Add 7 new editing functions with comprehensive error handling
- gdocs/docs_helpers.py: New helper module with utility functions for document operations
- README.md: Update Google Docs description to include full editing capabilities
- CLAUDE.md: Add comprehensive documentation for new editing features
- .mcp.json: Add MCP server configuration file
- .gitignore: Resolve merge conflicts from upstream sync
- uv.lock: Update dependencies from upstream merge
The implementation uses Google Docs API batchUpdate operations for atomic
document modifications and maintains compatibility with existing authentication
and error handling patterns.
All new tools follow the established @require_google_service decorator pattern
and include proper scope management for write operations.
