Single tool 'active_note' has a reasonable description (89 chars, within 10-1024 range) but critically lacks an input schema. The tool accepts no parameters (empty input object), yet this is not formally documented in a JSON Schema. Tool naming lacks action verb prefix (should be 'get_active_note'). No output schema documented. No error handling guidance. No parameter descriptions (N/A since no params, but reinforces incomplete specification). This server exposes a read-only Obsidian integration but fails basic definition quality standards.

• Rename 'active_note' to 'get_active_note' to follow verb_noun convention and signal retrieval semantics to LLMs.
• Add formal JSON Schema for input (even if empty: {"type": "object", "properties": {}, "required": []}) and output (document the text content structure).
• Expand description to 100-150 chars: 'Retrieve the path and full content of the currently active Obsidian note. Returns empty if no note is open. Use this to ground agent context in the user's current work.'
• Return structured error responses: {"error": "no_note_open", "message": "No note is currently active. Ask the user to open a note first."} instead of plain text.
• Document output schema in tool registration: {"type": "object", "properties": {"path": {"type": "string"}, "content": {"type": "string"}}, "required": ["path", "content"]}