Loading ToolBench
21 tools with consistent naming (verb_noun pattern) and complete JSON schemas. All tools have descriptions (avg 65 chars, baseline 194 chars) and input schemas with typed parameters. However, descriptions are generic and lack LLM-optimization guidance (when to use, prerequisites, what it returns). No output schemas documented. Parameter descriptions are minimal (avg 25 chars, baseline 72 chars). No error handling guidance, no recovery hints, no confirmation patterns for destructive ops. Missing tool annotations (readOnlyHint, destructiveHint). Schemas are present but sparse, many parameters lack constraints (enums, ranges, formats). No pagination guidance in list tools despite accepting cursor/per_page. Overall: functional but below production baseline.
Create a new cycle (sprint)
Create a new issue in a project
Add an external link to an issue
Create a relation between two issues
Delete a cycle
Delete an issue
Remove an external link from an issue
Descriptions are generic and lack LLM-optimization guidance. Most descriptions (avg 65 chars) are below the 194-char baseline and do not explain WHEN to use the tool, WHAT it returns, or dependencies. E.g., 'List all workspaces accessible to the authenticated user' does not hint at pagination, filtering, or expected response structure.
Parameter descriptions are minimal (avg 25 chars vs baseline 72 chars). Most parameters lack format guidance, constraints, or examples. E.g., 'Filter by priority' does not explain that enum values are [urgent, high, medium, low, none]; 'Pagination cursor' does not explain cursor format or how to use it.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 51 | <=2025-11-25 | v2 |
Get details of a specific cycle
Get details of a specific issue
Get details of a specific project
List all issues in a cycle
List all cycles (sprints) in a project
List activity/history for an issue
List all external links attached to an issue
List all relations for an issue
List issues in a project with optional filters
List all projects in the workspace
List all sub-issues of a parent issue
List all workspaces accessible to the authenticated user
Update an existing cycle
Update an existing issue
No output schemas documented. LLMs cannot plan downstream tool calls or extract required fields (e.g., issue_id from create_issue response). Baseline: 100% of A+ tools document return types.
Destructive tools (plane_delete_issue, plane_delete_cycle, plane_delete_issue_link) lack confirmation patterns or dry-run support. No error handling guidance. Agents cannot safely retry or recover from mistakes.
No tool annotations (readOnlyHint, destructiveHint, idempotentHint). LLMs cannot distinguish safe reads from destructive writes without explicit hints. Baseline: A+ tools use annotations to guide agent planning.