{"id":"31fbd5a4-a26f-4ca3-a719-36c688b4d1b8","slug":"clawhub-kokxi-qa-ai-prompt-strategy","name":"qa-ai-prompt-strategy","description":"根据不同的测试目标和上下文，选择最佳的提示词模式来驱动AI生成高质量的测试用例。当AI输出的测试用例质量不够好、太泛泛、或者深度不够时，问题往往不在AI而在提示词。此技能提供结构化提示词模板，注入前面步骤产出的分析结果，输出包含角色定义、输出格式规范和约束条件的优化提示词。⚠️ 作为工作流的必过步骤，不得跳过。 触发场景：怎么问AI、AI回答不好、换个方式问、提示词、提问模板、提示词优化、角色扮演、AI输出太浅需要更深时。 Use when the user asks about: choosing or optimizing the prompt that drives test case generation — role definition, output format constraints, and injection of prior analysis results.","canonicalUrl":"https://www.xpersona.co/agent/clawhub-kokxi-qa-ai-prompt-strategy","sourceUrl":"https://clawhub.ai/kokxi/qa-ai-prompt-strategy","homepage":"https://clawhub.ai/kokxi/skills/qa-ai-prompt-strategy","source":"CLAWHUB","vendor":{"slug":"clawhub","label":"Clawhub","url":"https://clawhub.ai/kokxi/skills/qa-ai-prompt-strategy"},"protocols":["OPENCLEW"],"capabilities":[],"trustScore":null,"trustConfidence":"unknown","artifactCount":0,"benchmarkCount":0,"lastRelease":"1.8.0","freshnessAt":"2026-10-11T07:15:28.430Z","freshnessLabel":"Oct 11, 2026","securityReviewed":true,"openapiReady":false,"stats":[{"label":"Trust score","value":"Unknown"},{"label":"Compatibility","value":"OpenClaw"},{"label":"Freshness","value":"Oct 11, 2026"},{"label":"Vendor","value":"Clawhub"},{"label":"Artifacts","value":"0"},{"label":"Benchmarks","value":"0"},{"label":"Last release","value":"1.8.0"}],"factsPreview":[{"factKey":"vendor","category":"vendor","label":"Vendor","value":"Clawhub","href":"https://clawhub.ai/kokxi/skills/qa-ai-prompt-strategy","sourceUrl":"https://clawhub.ai/kokxi/skills/qa-ai-prompt-strategy","sourceType":"profile","confidence":"medium","observedAt":"2026-10-11T07:15:28.443Z","isPublic":true},{"factKey":"protocols","category":"compatibility","label":"Protocol compatibility","value":"OpenClaw","href":"https://www.xpersona.co/api/v1/agents/clawhub-kokxi-qa-ai-prompt-strategy/contract","sourceUrl":"https://www.xpersona.co/api/v1/agents/clawhub-kokxi-qa-ai-prompt-strategy/contract","sourceType":"contract","confidence":"medium","observedAt":"2026-10-11T07:15:28.443Z","isPublic":true},{"factKey":"traction","category":"adoption","label":"Adoption signal","value":"1.1K downloads","href":"https://clawhub.ai/kokxi/qa-ai-prompt-strategy","sourceUrl":"https://clawhub.ai/kokxi/qa-ai-prompt-strategy","sourceType":"profile","confidence":"medium","observedAt":"2026-10-11T07:15:28.443Z","isPublic":true},{"factKey":"latest_release","category":"release","label":"Latest release","value":"1.8.0","href":"https://clawhub.ai/kokxi/qa-ai-prompt-strategy","sourceUrl":"https://clawhub.ai/kokxi/qa-ai-prompt-strategy","sourceType":"release","confidence":"medium","observedAt":"2026-09-29T04:25:13.897Z","isPublic":true},{"factKey":"handshake_status","category":"security","label":"Handshake status","value":"UNKNOWN","href":"https://www.xpersona.co/api/v1/agents/clawhub-kokxi-qa-ai-prompt-strategy/trust","sourceUrl":"https://www.xpersona.co/api/v1/agents/clawhub-kokxi-qa-ai-prompt-strategy/trust","sourceType":"trust","confidence":"medium","observedAt":null,"isPublic":true}],"highlights":["1.1K downloads","Trust evidence available"],"agentCard":{"name":"qa-ai-prompt-strategy","description":"根据不同的测试目标和上下文，选择最佳的提示词模式来驱动AI生成高质量的测试用例。当AI输出的测试用例质量不够好、太泛泛、或者深度不够时，问题往往不在AI而在提示词。此技能提供结构化提示词模板，注入前面步骤产出的分析结果，输出包含角色定义、输出格式规范和约束条件的优化提示词。⚠️ 作为工作流的必过步骤，不得跳过。 触发场景：怎么问AI、AI回答不好、换个方式问、提示词、提问模板、提示词优化、角色扮演、AI输出太浅需要更深时。 Use when the user asks about: choosing or optimizing the prompt that drives test case generation — role definition, output format constraints, and injection of prior analysis results.","source":"CLAWHUB","sourceId":"clawhub:s170jw3s1atcj5jwhqb4r7v7eh8912kp:qa-ai-prompt-strategy","homepage":"https://clawhub.ai/kokxi/skills/qa-ai-prompt-strategy","repository":"https://clawhub.ai/kokxi/qa-ai-prompt-strategy","documentation":"https://www.xpersona.co/agent/clawhub-kokxi-qa-ai-prompt-strategy","protocols":["OPENCLEW"],"examples":[{"kind":"example","language":"text","snippet":"1. 角色定义：你是[领域]资深测试专家\n2. 输出数量：生成[N]条测试用例（N = 需求数量 × 5）\n3. 覆盖维度：必须覆盖以下维度\n   - 功能测试：[具体功能点]\n   - 异常测试：[异常场景类型]\n   - 边界测试：[边界条件类型]\n   - 并发测试：[并发场景]\n   - 安全测试：[安全风险点]\n   - 性能测试：[性能指标]\n4. 输出格式：Markdown表格，包含需求ID和风险ID\n5. 质量要求：每条用例必须可执行、可验证"},{"kind":"example","language":"text","snippet":"1. 数量约束：国产模型对\"生成 N 条用例\"偶有缩水 → 提示词写\"不少于 N 条\"，并加\"如果不足，补充边界和异常场景\"\n2. 格式约束：多数国产模型默认带 Markdown 代码块 → 需显式声明\"直接输出表格，不要代码块包裹\"\n3. 中文表达：国产模型中文流畅但对\"用例标题\"格式理解差异大 → 给出 1 条示例（如 TC_API_LOGIN_001 正确凭证登录成功）\n4. 编号规则：部分模型会自作主张改编号 → 明确\"严格使用 TC_{模块缩写}_{功能缩写}_{序号}，不得改动格式\"\n5. 深度控制：国产模型倾向少而浅 → 用深度量化（简单×5/中等×10/复杂×15）约束用例数量级"}]}}