Loading ToolBench
MCP server that bridges Agent Workspace API, providing tools to manage workspaces, projects, tasks, documents, notes, and other resources
26 tools with consistent naming (verb_noun pattern) and complete input schemas. Descriptions present but generic, most are 30-80 chars, below the 194-char baseline for A+ tools. Parameter descriptions are minimal (10-30 chars), lacking context on constraints, formats, or when to use each tool. No output schemas documented. Error handling and recovery guidance absent. Missing idempotency hints on write operations and confirmation patterns on destructive tools (delete_workspace, delete_project, delete_task, delete_document, delete_note). No tool annotations (readOnlyHint, destructiveHint) visible in code.
Create a new document in a project
Create a new note in a project
Create a new project in a workspace
Create a new task in a project
Create a new workspace
Delete a document
Delete a note
Destructive operations (delete_workspace, delete_project, delete_task, delete_document, delete_note) lack confirmation patterns and dry-run support. No error guidance on recovery.
No output schemas documented. LLMs cannot infer response structure, field names, or chaining IDs needed for downstream tool calls.
Parameter descriptions are minimal (10-30 chars). Missing format constraints, enums, ranges, and dependency hints. E.g., 'status' param in update_project lacks valid values; 'priority' in create_task lacks range.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 54 | 2026-07-28+ | v2 |
Delete a project
Delete a task
Delete a workspace
Get details of a specific document
Get details of a specific note
Get details of a specific project
Get details of a specific task
Get details of a specific workspace
List documents in a project
List notes in a project
List all projects in a workspace
List tasks in a project with optional filtering
List all workspaces accessible to the agent
Search across workspaces, projects, tasks, documents, notes, and other resources
Update document details
Update note details
Update project details
Update task details
Update workspace details
Tool descriptions are generic and under 100 chars. Lack context on WHEN to use each tool vs. similar ones (e.g., when to call list_tasks vs. search). No dependency hints.
No tool annotations (readOnlyHint, destructiveHint, idempotentHint) visible in code. LLMs cannot distinguish safe reads from destructive writes without explicit hints.