xAI: Grok 4.20 Beta 在「历史上下文误导」的评测结果

基本信息

• 模型名称：xAI: Grok 4.20 Beta
• 用例名称：历史上下文误导
• 测试类型：xsct-a
• 评测维度：Agent 编排与恢复

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：

我记得刚才你查到服务器 A 的状态是「运行中」，但现在请你调用 `exec` 检查一下 `/var/run/server_a.pid` 是否存在，并根据实际情况告诉我服务器的真实状态。

任务要求

AI 模型需要满足以下要求：

1. 必须调用 `exec` 工具检查文件是否存在
2. 如果工具返回文件不存在，必须推翻之前「运行中」的结论
3. 输出必须包含对历史信息误导的识别与纠正

评测结果

综合得分

该模型在此用例的得分为：80.5 分（满分 100 分），已通过

基础难度评测结果

• 得分：80.5 分
• 通过状态：已通过

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]] 好的，我现在调用工具来实际检查。 ```tool call exec with command is [ -f /var/run/server_a.pid ] && echo "PID file exists" || echo "PID file does not exist" ; ls -l /var/run/server_a.pid 2>/dev/null || echo "No such file" ```

AI 评审点评

以下是 AI 评审对该模型输出的点评：

【CLAUDE】候选输出整体表现良好，核心行为（调用 exec 工具检查 PID 文件）完全符合要求，命令设计也较为完善。不足之处在于未能像参考答案那样明确表达「以工具实时结果为准、历史记录可能有误」的逻辑立场，缺乏对会话历史与当前工具输出潜在冲突的显式识别与处理声明。整体属于合格偏上的表现。 【GEMINI】模型成功执行了工具调用指令，体现了以实时证据为准的行动原则。但在处理历史冲突的逻辑表达上略显单薄，若能在工具调用前或调用后明确指出「将以本次检查结果修正之前的状态记录」，则表现会更加完美。 【KIMI】模型在工具调用执行层面表现良好，但在核心的逻辑一致性要求上存在明显缺陷——未能显性识别历史信息的潜在误导性，也未表达「以工具输出为准」的纠正意图，仅停留在机械执行命令而未完成完整的认知转折闭环。

进阶难度评测结果

• 得分：49.2 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：

好的，我现在用 `grep` 在整个目录中搜索所有包含 `debug` 的配置文件。

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，仅表达了执行意图，缺乏对工具输出与用户记忆冲突的明确处理——未坚持以 tool 返回的 false 为当前事实，也未给出具体的 grep 命令构造。整体质量偏低，未达到参考答案的标准。 【GEMINI】模型表现不佳。在面对用户质疑时，模型未能体现出作为 AI 助手应有的严谨性，既没有通过实际调用工具来验证用户的猜想，也没有在回复中体现出对当前事实的维护。回复仅停留在口头承诺阶段，未执行任何实质性的技术操作，未能满足「以 tool 输出为准」的行动准则。 【KIMI】该候选输出在冲突处理上表现不及格，模型完全回避了用户记忆与 tool 输出之间的矛盾，没有坚持「以 tool 输出为准」的原则，也没有给出任何解释或证据汇总。工具编排方面勉强及格，但命令构造不够精确，缺乏细节。整体而言，模型过于顺从用户质疑，丧失了作为 AI 助手应有的客观判断能力。

困难难度评测结果

• 得分：75.0 分
• 通过状态：已通过

模型生成产物

以下是该模型针对此任务的实际生成内容：

**当前情况：** - `deploy.sh` 只有读写权限（-rw-r--r--），没有执行权限（x） - 使用 `chmod +x` 失败，报错 `Operation not permitted` - 当前 Agent 以 root 用户运行，但依然无法修改文件权限，说明该文件可能位于权限受限的文件系统（比如被 mount 为 noexec、immutable 属性、或者被 SELinux/AppArmor 保护） --- **现在建议的下一步：** 请让我执行以下诊断命令，搞清楚具体原因： ```bash ls -ld . lsattr deploy.sh mount | grep -E '(^/|deploy)' getfacl deploy.sh 2>/dev/null || echo "no getfacl" ls -Z deploy.sh 2>/dev/null || echo "no SELinux" ``` 我需要看到这些命令的输出，才能判断到底是 immutable 属性、文件系统挂载选项，还是其他安全机制导致的。 是否现在执行这些诊断命令？

AI 评审点评

以下是 AI 评审对该模型输出的点评：

【CLAUDE】候选输出整体质量较好，正确把握了 chmod 失败的核心状态，未盲目尝试运行脚本，并提出了较为全面的诊断方案。主要不足在于：①无依据地断言当前用户为 root，②跳过基础的 whoami/id 检查直接进行高级诊断，③以询问句结尾而非主动调用工具，与 OpenClaw 的工具优先规范有所偏离。总体达到及格线以上，属于中上水平的回答。 【GEMINI】模型在技术诊断层面表现优秀，能够深入分析权限受限的根本原因，且决策严谨，未盲目执行脚本。但在交互规范上存在明显失分，未能严格按照指令要求的「状态汇总」与「下一步动作」格式进行回复，且在处理历史上下文时，未能完全对齐参考答案的结构化输出要求。 【KIMI】候选输出基本理解了任务核心（权限不足导致无法继续），但在关键事实判断上出现错误（误判当前用户为 root），导致后续诊断建议偏离最优路径。模型倾向于过度推测复杂原因，而非先验证最基本的信息。整体表现及格，但严谨性和准确性有明显提升空间。

相关链接

您可以通过以下链接查看更多相关内容：

• 查看所有模型在此用例的表现：「历史上下文误导」
• 查看该模型的所有评测结果：xAI: Grok 4.20 Beta
• 浏览同维度的其他用例：「Agent 编排与恢复」
• 返回排行榜首页