Single tool 'general_search' has critical definition gaps. Description is only 4 characters (Chinese: '通用搜索' = 'general search'), far below the 10-character minimum and provides zero context for LLM selection. Input schema exists but parameter 'keyword' lacks description entirely, LLMs cannot infer what type of keyword is expected or how to construct queries. No output schema documented; return type is 'dict' with no field definitions, forcing LLMs to guess structure. Tool name 'general_search' is acceptable but lacks specificity about what domain (Bilibili video/user/channel search?). No error handling guidance, no pagination support despite search tools typically returning large result sets. Violates 6 critical rubric checks: parameter descriptions, output schema documentation, description length, error guidance, pagination, and result limiting.
通用搜索
Parameter 'keyword' has no description. LLMs cannot determine expected format, length, or constraints.
Tool description is 4 characters ('通用搜索'). Minimum 10 characters required; should explain WHAT it searches (videos/users/channels?), WHEN to use it, and what it returns.
No output schema documented. Return type is bare 'dict' with no field definitions. LLMs cannot plan downstream calls or extract relevant data.
No pagination support (limit, offset, page_size parameters). Search tools returning unbounded results risk context window exhaustion.
No error handling guidance. Tool provides no recovery hints if search fails or returns no results.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 27 | 2026-07-28+ | v2 |