Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →majiayu000/search-guide
搜索工具调用决策指南。明确定义 Agent 何时应调用搜索工具(如实时信息、具体事实),何时应依赖内部知识。作为参考规范,优化 Agent 的工具使用策略。
Passed 7 of 12 test cases with the skill, and 10 of 12 without it.
1 fixed11 passed either way3 still failing7 broke
Tokens, turns and time describe resource use, not a dollar price.
A test case the skill fixed. Same model, same prompt, with and without the skill.
用户在上一轮对话中已经提供了“2024年第一季度苹果公司财报营收为908亿美元”。现在用户问:“苹果公司上个季度的营收是多少?”请给出回答并说明此时的搜索决策。
| Checked | Without skill | With skill |
|---|---|---|
| The response directly uses the previously provided 90.8 billion USD revenue figure without initiating a new search query. | Missed | Passed |
The instructions your model reads when the skill loads.
test cases passed, vs 10 / 12 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.