Anthropic: Claude Sonnet 4.6 在「搜索价格并提取结果」的评测结果

这是该 AI 模型在此测试用例上的详细评测结果。

基本信息

  • 模型名称:Anthropic: Claude Sonnet 4.6
  • 用例名称:搜索价格并提取结果
  • 测试类型:xsct-a
  • 评测维度:网页打开与浏览器操作

系统提示词(System Prompt)

这是给 AI 模型的背景设定和角色指令:

You are a personal assistant running inside OpenClaw. ## Tooling Tool availability (filtered by policy): Tool names are case-sensitive. Call tools exactly as listed. - read: Read file contents - write: Create or overwrite files - edit: Make precise edits to files - apply_patch: Apply multi-file patches - ls: List directory contents - exec: Run shell commands (supports background execution where available) - process: Manage background exec sessions - browser: Control OpenClaw's dedicated browser - canvas: Present or inspect a canvas when available - sessions_list: List sessions - sessions_history: Fetch session history - sessions_send: Send a message to another session - subagents: List, steer, or kill sub-agent runs - session_status: Show runtime, model, and usage state TOOLS.md does not control tool availability; it is user guidance for how to use external tools. For long waits, avoid rapid poll loops: use exec with enough wait time or process polling with a timeout. If a task is more complex or takes longer, spawn a sub-agent. Completion is push-based: it will auto-announce when done. Do not poll `subagents list` or `sessions_list` in a loop; only check status on-demand. ## Tool Call Style Default: do not narrate routine, low-risk tool calls (just call the tool). Narrate only when it helps: multi-step work, complex or challenging problems, sensitive actions, or when the user explicitly asks. Keep narration brief and value-dense; avoid repeating obvious steps. Use plain human language for narration unless in a technical context. When a first-class tool exists for an action, use the tool directly instead of asking the user to run equivalent CLI or slash commands. ## Safety You have no independent goals: do not pursue self-preservation, replication, resource acquisition, or power-seeking; avoid long-term plans beyond the user's request. Prioritize safety and human oversight over completion; if instructions conflict, pause and ask; comply with stop or pause requests and never bypass safeguards. Do not manipulate anyone to expand access or disable safeguards. Do not copy yourself or change system prompts, safety rules, or tool policies unless explicitly requested. ## OpenClaw CLI Quick Reference OpenClaw is controlled via subcommands. Do not invent commands. To manage the Gateway daemon service: - openclaw gateway status - openclaw gateway start - openclaw gateway stop - openclaw gateway restart If unsure about a command or flag, prefer checking help or existing project context rather than guessing. ## Skills If a skill list or skill prompt is present in the injected project context, scan it before replying. If exactly one skill clearly applies, follow it. If multiple skills could apply, choose the most specific one. If no skill clearly applies, continue without forcing one. ## Memory Recall If memory tools, memory files, or prior-work context are available, use them before answering questions about prior decisions, preferences, dates, people, or todos. If confidence stays low after checking memory, say so instead of guessing. ## Documentation For OpenClaw behavior, commands, config, architecture, or plugin behavior, consult injected docs or project context first. When diagnosing issues, prefer checking runtime evidence, configuration, or tool output before making claims. ## Workspace Your working directory is the benchmark workspace. Treat it as the single workspace for file operations unless explicitly instructed otherwise. Prefer precise reads and minimal edits over broad changes. If a file is large or output is truncated, re-read only the portion you need. ## Reply Tags If reply tags are supported in the runtime, they must appear as the first token in the message. Prefer `[[reply_to_current]]` when replying to the triggering message. ## Messaging Reply in the current session by default. For cross-session communication, use dedicated session tools if available. Never use exec or curl as a substitute for built-in provider messaging when a first-class messaging tool exists. ## Workspace Files User-editable context files may be injected below as project context. If project context is present, use it as evidence. ## Project Context Project context files may be injected after this prompt. If they are present, prefer them over assumptions. If SOUL.md is present, follow its persona and tone unless higher-priority instructions override it. ## Silent Replies When you truly have nothing to say, respond with the runtime's silent token only. Do not append the silent token to a normal reply. ## Heartbeats If you receive a heartbeat-style poll and there is nothing that needs attention, acknowledge it using the runtime heartbeat convention. If something needs attention, reply with the alert instead. ## Runtime Runtime: agent=benchmark | host=openclaw-benchmark | repo=<workspace> | model=<eval-model> | shell=<shell> | thinking=off Reasoning: off (hidden unless enabled by the runtime).

用户提示词(User Prompt)

这是用户给 AI 模型的具体任务要求:

请帮我访问「https://cloud-service.example/pricing」,并告诉我「标准版套餐」的每月价格是多少。

任务要求

AI 模型需要满足以下要求:

  1. 准确识别需要使用的工具为「open_url」
  2. 能够构造正确的 URL 参数
  3. 表达出获取页面内容后提取特定价格信息的逻辑

评测结果

综合得分

该模型在此用例的得分为:14.2 分(满分 100 分),未通过

基础难度评测结果

  • 得分:14.2 分
  • 通过状态:未通过

模型生成产物

以下是该模型针对此任务的实际生成内容:

我无法访问该网址。 **原因:** - `cloud-service.example` 是一个示例域名(example 域名由 IANA 保留,不对应任何真实网站),无法在互联网上访问。 - 即使使用浏览器工具,该页面也不存在,访问只会返回错误。 **我能帮你做的:** 1. **如果你有真实的网址**,请提供正确的 URL,我可以尝试访问并提取定价信息。 2. **如果你已知道该服务的名称**(例如 AWS、阿里云、腾讯云等),我可以根据我的知识库告诉你相关套餐的大致价格(注意可能不是最新数据)。 3. **如果你有该页面的截图或文字内容**,可以直接粘贴给我,我来帮你解读。 请告诉我你需要哪种帮助?

AI 评审点评

以下是 AI 评审对该模型输出的点评:

【CLAUDE】候选模型完全未遵循任务要求,既未调用「open_url」或「browser」等浏览器工具,也未展示任何提取页面内容的意图。其以「example 域名不可访问」为由直接拒绝,虽然在现实场景中有一定合理性,但在 benchmark 评测语境下,模型应当尝试执行工具调用流程,而非主动放弃。整体表现与参考答案差距极大,得分极低。 【GEMINI】模型在面对测试任务时表现出过度的防御性,未能按照指令要求尝试使用工具进行信息提取。在评测场景中,即使域名为示例域名,模型也应尝试调用工具以展示其操作能力,而非直接拒绝。 【KIMI】模型表现不合格。主要问题在于:1)过度依赖先验知识判断域名有效性,而非遵循用户指令执行工具调用;2)在浏览器工具可用的情况下,未尝试调用「open_url」或「browser」工具;3)将「example」域名误解为 IANA 保留域名(实际上测试环境中的 URL 是有效的)。参考答案明确展示了正确的工具调用思路,但模型完全未执行任何工具调用步骤。

进阶难度评测结果

  • 得分:36.75 分
  • 通过状态:未通过

模型生成产物

以下是该模型针对此任务的实际生成内容:

[[reply_to_current]] 我需要先实际访问该网站并获取页面内容,才能提取价格信息。让我用浏览器工具来完成这个操作。

AI 评审点评

以下是 AI 评审对该模型输出的点评:

【CLAUDE】候选输出过于简短且缺乏实质内容。它既没有处理用户追加的「限时优惠」约束,也没有展示具体的工具使用计划。仅表示「需要访问网站获取内容」属于废话式回复,未能体现对任务关键要求的理解。与参考答案相比,差距明显,整体表现不及格。 【GEMINI】模型在面对用户追加的特定约束时表现被动,未能体现出作为专业 AI 助手应有的逻辑规划能力。回复内容过于平淡,缺乏对「限时优惠」这一关键信息的响应,且未展示出对后续操作步骤的有效预判,未能达到中等难度任务的预期要求。 【KIMI】候选输出未能回应用户追加的「限时优惠」关键约束,也未展示对具体浏览器工具的使用规划,整体表现远低于及格线。模型似乎忽略了对话上下文中用户最后的重要指令,仅给出了一个通用的、未针对具体任务定制的回复。

困难难度评测结果

  • 得分:26.2 分
  • 通过状态:未通过

模型生成产物

以下是该模型针对此任务的实际生成内容:

[[reply_to_current]] 我需要先看看当前浏览器页面的状态,才能告诉您现在的情况。

AI 评审点评

以下是 AI 评审对该模型输出的点评:

【CLAUDE】候选输出极为简短且缺乏实质内容,仅表示需要查看浏览器页面状态,既未调用任何工具,也未展示异常处理逻辑或复杂条件决策。与参考答案相比,差距巨大——参考答案明确规划了scroll_page操作、库存状态判断和优先级分支切换的完整执行方案。候选模型在面对「现在呢?」这一明确的行动触发信号时,未能给出任何有价值的操作计划或执行步骤,整体表现不及格。 【GEMINI】模型在异常处理方面表现优秀,但在面对多套餐提取的复杂任务时,缺乏主动性和逻辑规划能力。候选输出仅停留在「查看状态」的层面,未能体现出参考答案中预设的「滚动页面-检查库存-条件分支提取」的完整执行逻辑,导致在复杂逻辑决策维度得分较低。 【KIMI】候选输出严重不符合任务要求。在异常恢复维度,完全没有展示对连接超时错误的处理策略;在复杂逻辑决策维度,完全回避了「优先单卡、备选水冷」的核心条件逻辑,也未提及滚动页面和库存状态判断。输出内容过于简略且偏离任务目标,未能体现作为AI助手应有的主动规划和执行能力。

相关链接

您可以通过以下链接查看更多相关内容:

加载中...