Loading ToolBench
Critical failures across naming, descriptions, and schemas. Tool 'authenticate' has a malformed inputSchema (properties.topic.url instead of properties.url). 'post_thread' lacks descriptions for 3 of 4 parameters. Both tools have vague, short descriptions (under 50 chars). No output schemas documented. No error recovery guidance. This server would not pass code review.
Provide authentication URL to get token for Threads API.
Create a new post to Threads
authenticate tool has malformed inputSchema: properties.topic.url instead of properties.url. Schema is invalid JSON Schema and will fail validation.
post_thread missing parameter descriptions for 'carousel' and 'photosIncluded'. LLMs cannot infer what these boolean flags control without explicit descriptions.
Tool descriptions are too short and vague. 'Create a new post to Threads' (28 chars) and 'Provide authentication URL to get token for Threads API.' (57 chars) lack context on WHEN to use, WHAT happens, and prerequisites. Baseline is 194 chars.
No output schemas documented. LLMs cannot plan downstream calls or extract return values. Both tools return TextContent but structure is not declared in tool definitions.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 32 | <=2025-11-25 | v2 |
Error handling returns raw McpError or unstructured text. No recovery guidance (e.g., 'Try search_users() first'). Errors like 'Invalid parameters' do not tell the LLM what to do next.