Loading ToolBench
MCP server for UpNote. Lets Claude and other MCP clients read your notes and create new ones.
UpNote MCP has 10 well-named tools with clear verb-noun patterns (create_, list_, search_, get_, open_). Descriptions are present and reasonably detailed (avg ~120 chars), explaining WHAT each tool does and WHEN to use it. Input schemas are complete with proper JSON Schema types and descriptions for all parameters. However, output schemas are not documented, LLMs cannot predict what fields to expect from responses. Error handling is minimal; no recovery guidance or actionable error messages visible. No tool annotations (readOnlyHint/destructiveHint) despite clear read/write distinction. Parameter descriptions could be more prescriptive about formats and constraints. Overall solid foundation but lacks production-grade polish.
Create a new note in UpNote. Use this to save a summary, plan, snippet or session output. If no notebook is given it goes to "Claude Notes". Content is Markdown. Cannot set tags and cannot edit an existing note.
Create a new notebook in UpNote.
Get the full text of one note by id.
List every UpNote notebook with its live note count.
List the notes in one notebook, newest first.
List all tags in UpNote.
Open a note in the UpNote app by id.
Output schemas not documented. LLMs cannot predict response structure (fields, types, pagination). Forces agents to guess what data is available after each call.
No tool annotations (readOnlyHint/destructiveHint). Despite clear read/write distinction in Risk field, MCP annotations are absent. Agents cannot distinguish safe vs. destructive operations.
Minimal error handling. No recovery guidance, actionable error messages, or error classification (retryable vs. fatal). Agents cannot self-correct on failures.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | B | 70 | 2026-07-28+ | v2 |
Open a notebook in the UpNote app by title.
The most recently updated notes across the whole library.
Full text search across note titles and bodies. Optionally scoped to one notebook.
Parameter descriptions lack format/constraint details. E.g., 'notebook' accepts string but no guidance on case sensitivity, partial matching, or error handling if notebook not found.
No pagination documentation. list_notes and list_tags accept 'limit' but no mention of offset, cursor, or total count. Large result sets could exhaust context.