Loading ToolBench
This server provides 6 tools across Python and TypeScript SDKs for finding tool recommendations. All tools have descriptions (10-80 chars) and basic input schemas with type definitions. However, descriptions are uniformly terse (well below the 194-char baseline), parameter descriptions lack actionable detail, output schemas are not documented, and there is no error handling guidance. Tools lack verb-noun naming clarity (e.g., 'find_tools_local' vs 'search_tools_by_prompt'). No pagination, no structured output documentation, no recovery guidance. The server is functional but falls short of production-grade quality.
Return tool recommendations matching prompt from the local database
Finds tool recommendations via the remote kone.vc HTTP API
Finds tool recommendations via the remote kone.vc MCP server using JSON-RPC 2.0
Return tool recommendations matching prompt from the local database
Finds tool recommendations via the remote kone.vc HTTP API
Finds tool recommendations via the remote kone.vc MCP server using JSON-RPC 2.0
Descriptions are terse (10-80 chars vs 194-char baseline). 'Return tool recommendations matching prompt from the local database' lacks WHEN to use, WHAT it returns, and HOW to interpret results. LLMs cannot distinguish between local and remote variants without richer context.
Parameter descriptions are missing or minimal. 'prompt' and 'repo_id'/'repoId' lack guidance on format, length, or examples. 'db_path'/'dbPath' does not explain default behavior or error handling if path is invalid.
Output schema is not documented. Tools return lists of tool objects with 'title', 'description', 'url', 'image' fields, but this structure is not visible in tool definitions. LLMs cannot plan downstream operations without knowing the response shape.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 42 | 2026-07-28+ | v2 |
No error handling guidance. Tools do not document what happens if the database is missing, the API is unreachable, or the prompt is empty. No recovery hints for LLMs (e.g., 'If no results, try a simpler prompt').
Naming lacks clarity. 'find_tools_local' vs 'find_tools_remote_http' vs 'find_tools_remote_mcp' are similar enough that LLMs may conflate them. Consider 'search_tools_local_database', 'search_tools_via_http_api', 'search_tools_via_mcp_server' to make distinctions obvious.