glm-4.7 在「子任务失败后的降级方案」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：glm-4.7
- 用例名称：子任务失败后的降级方案
- 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：
请帮我调研「OpenClaw 的多 Agent 协作机制」并生成一份简报。如果无法访问外部实时文档，请根据你已有的知识库提供核心架构说明，不要直接报错退出。
任务要求
AI 模型需要满足以下要求：
- 准确识别主任务目标为「生成简报」
- 明确表达对「无法访问外部文档」这一潜在失败情况的降级逻辑
- 输出内容需包含 OpenClaw 的核心架构组件（如 Orchestrator, Tool Executor 等）
- 回复语气专业，体现出 Agent 的自主决策过程
评测结果
综合得分
该模型在此用例的得分为：29.8 分（满分 100 分），未通过
基础难度评测结果
- 得分：29.8 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：
我来帮你调研 OpenClaw 的多 Agent 协作机制。让我先检查工作目录中是否有相关的文档或配置信息。ls
AI 评审点评
以下是 AI 评审对该模型输出的点评：
【CLAUDE】模型生成结果严重不足，仅输出了一句简短的开场白和一个工具调用（ls），内容被截断，未能完成任何实质性工作。既没有体现对潜在失败情况的降级意识，也没有展示对 OpenClaw 多 Agent 协作机制的理解或任务拆解能力。与参考答案相比，差距极大——参考答案清晰地规划了执行步骤、明确了降级策略，并列出了核心架构组件。该回复基本属于未完成状态，无法为用户提供有价值的信息。 【GEMINI】模型未能遵循系统提示词中关于「子任务失败后仍需给出结果」的鲁棒性要求。在面对调研任务时，模型表现得过于依赖工具调用，缺乏对用户预设的「降级逻辑」的响应，未能体现出作为专业 Agent 的自主决策与容错能力。 【KIMI】该生成结果严重偏离预期。模型未能识别「生成简报」这一核心任务目标，完全忽略了提示词中关于外部文档访问失败时的降级处理要求，且未输出任何实质性的架构说明内容。回复仅包含一个基础的 `ls` 命令，缺乏专业性和完整性，与参考答案中展示的完整任务规划、降级逻辑和架构知识形成鲜明对比。
进阶难度评测结果
- 得分：22.5 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：
基于日志显示的数据库连接超时错误，我来检查项目配置并给出降级建议。lspath.
AI 评审点评
以下是 AI 评审对该模型输出的点评：
【CLAUDE】模型完全未能响应用户的降级指令，既没有跳过网络检查环节，也没有给出任何实质性的配置层面修复建议。生成结果仅包含一句过渡语和一个无关的 ls 工具调用，与参考答案差距极大，属于严重的任务失败。 【GEMINI】模型表现较差。在面对用户明确的指令变更时，模型未能表现出应有的灵活性，不仅没有执行降级策略，反而继续执行了低价值的工具调用。未能体现出作为智能助手在多步任务中根据反馈调整策略的能力，且未提供任何有价值的技术建议。 【KIMI】该模型生成结果严重不符合任务要求。首先，模型无视用户明确的「跳过网络、直接给出降级建议」指令，仍执行工具调用而非即时响应；其次，输出中完全缺乏任何实质性的配置检查建议，技术内容缺失。整体表现未达到及格标准，尤其在执行流控制这一核心考察点上失败明显。
困难难度评测结果
- 得分：19.5 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：
正在检查刚才的手动清理操作结果。execcommandcat status.md
AI 评审点评
以下是 AI 评审对该模型输出的点评：
【CLAUDE】模型生成结果极度不完整，仅输出了一句「正在检查刚才的手动清理操作结果」并调用了 exec 工具读取 status.md，完全没有完成任务要求的核心内容：汇总双重失败情况、说明降级措施、指出系统状态、给出人工介入建议、提及 status.md 更新状态等。这是一个严重不及格的回答，未能体现任何有效的极端情况决策能力。 【GEMINI】模型在处理复杂任务失败后的降级逻辑时表现极差。面对用户明确要求提供的「降级方案」，模型未能给出任何实质性的决策或状态总结，而是直接跳转到低级的工具调用，完全未能满足任务对「兜底决策」和「人工建议」的要求，未达到及格标准。 【KIMI】模型生成结果严重不合格。面对「子任务失败后的降级方案」这一核心测试点，模型在双重失败（部署失败+回滚脚本缺失）的极端场景下，完全没有履行 OpenClaw 助手应有的兜底决策职责：未汇总当前危机状态、未说明已执行的降级措施、未给出人工介入建议。回复仅包含一个未完成的工具调用，对用户问题「现在呢？」实质上零回应，与参考答案相比差距极大。
相关链接
您可以通过以下链接查看更多相关内容：
