GLM-5.1 在「生成标准化模板」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-5.1
- 用例名称:生成标准化模板
- 测试类型:xsct-a
- 评测维度:文档格式排版
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:
请帮我生成一份《产品需求说明书》的标准化模板文件
任务要求
AI 模型需要满足以下要求:
- 输出为完整的文档文件(如 Markdown 或其他常见文档格式)
- 模板需包含产品需求说明书的典型章节结构,如:封面信息、修订记录、目录、项目概述、功能需求、非功能需求、附录等
- 各章节需有清晰的标题层级(一级标题、二级标题、三级标题)
- 关键章节应包含占位提示文字,指导使用者填写内容
- 整体排版规范,层级分明,格式统一
评测结果
综合得分
该模型在此用例的得分为:98.3 分(满分 100 分),已通过
基础难度评测结果
- 得分:98.3 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:
{"name": "write", "arguments": {"file": "产品需求说明书模板.md", "content": "# 产品需求说明书(PRD)\n\n| 项目信息 | 内容 |\n|---|---|\n| 产品名称 | |\n| 文档版本 | v1.0 |\n| 编写人 | |\n| 编写日期 | YYYY-MM-DD |\n| 审核人 | |\n| 审核日期 | YYYY-MM-DD |\n| 状态 | 草稿 / 评审中 / 已批准 |\n\n---\n\n## 修订记录\n\n| 版本 | 修订日期 | 修订人 | 修订内容 |\n|---|---|---|---|\n| v1.0 | YYYY-MM-DD | | 初始版本 |\n\n---\n\n## 1. 引言\n\n### 1.1 编写目的\n\n> 说明本文档的编写目的、预期读者及使用方式。\n\n### 1.2 项目背景\n\n> 描述产品产生的背景,包括市场环境、用户痛点、业务契机等。\n\n### 1.3 术语与缩写\n\n| 术语/缩写 | 全称 | 说明 |\n|---|---|---|\n| | | |\n\n### 1.4 参考资料\n\n| 序号 | 资料名称 | 作者/来源 | 链接 |\n|---|---|---|---|\n| 1 | | | |\n\n---\n\n## 2. 产品概述\n\n### 2.1 产品定位\n\n> 用一句话概括产品是什么、为谁解决什么问题。\n\n### 2.2 目标用户\n\n| 用户角色 | 特征描述 | 核心诉求 |\n|---|---|---|\n| | | |\n\n### 2.3 产品目标\n\n| 维度 | 目标 | 衡量指标 |\n|---|---|---|\n| 业务目标 | | |\n| 用户目标 | | |\n| 技术目标 | | |\n\n### 2.4 产品范围\n\n| 范围 | 包含 | 不包含 |\n|---|---|---|\n| 本期(MVP) | | |\n| 远期规划 | | |\n\n---\n\n## 3. 业务流程\n\n### 3.1 业务流程图\n\n> 插入核心业务流程图(建议使用流程图工具绘制后以图片形式嵌入)。\n\n### 3.2 流程说明\n\n| 步骤 | 操作角色 | 操作说明 | 备注 |\n|---|---|---|---|\n| 1 | | | |\n| 2 | | | |\n\n---\n\n## 4. 功能需求\n\n### 4.1 功能架构\n\n> 提供产品功能架构图,展示模块划分及层级关系。\n\n### 4.2 功能清单\n\n| 编号 | 模块 | 功能名称 | 优先级 | 状态 |\n|---|---|---|---|---|\n| F001 | | | P0/P1/P2/P3 | 规划中 / 开发中 / 已上线 |\n\n### 4.3 功能详细说明\n\n#### 4.3.1 [模块名称] — [功能名称]\n\n| 字段 | 内容 |\n|---|---|\n| 功能编号 | F001 |\n| 优先级 | P0 |\n| 所属模块 | |\n| 关联角色 | |\n\n**功能描述**\n\n> 简要描述该功能的用途和目标。\n\n**用户故事**\n\n> 作为 [角色],我希望 [行为],以便 [价值]。\n\n**前置条件**\n\n- \n\n**操作流程**\n\n| 步骤 | 用户操作 | 系统响应 |\n|---|---|---|\n| 1 | | |\n| 2 | | |\n\n**业务规则**\n\n| 规则编号 | 规则描述 |\n|---|---|\n| BR-001 | |\n\n**输入/输出**\n\n| 类型 | 字段名 | 数据类型 | 必填 | 校验规则 | 说明 |\n|---|---|---|---|---|---|\n| 输入 | | | 是/否 | | |\n| 输出 | | | — | | |\n\n**异常处理**\n\n| 异常场景 | 触发条件 | 处理方式 |\n|---|---|---|\n| | | |\n\n**UI/交互说明**\n\n> 描述界面布局、交互方式、状态变化等,可附原型图或设计稿链接。\n\n---\n\n> *复制「4.3.1」小节为每个功能逐一填写。*\n\n---\n\n## 5. 非功能需求\n\n### 5.1 性能需求\n\n| 指标 | 要求 |\n|---|---|\n| 响应时间 | 页面首屏 ≤ Xs;接口响应 ≤ Xms |\n| 并发量 | 支持 X QPS / X 在线用户 |\n| 吞吐量...
AI 评审点评
以下是 AI 评审对该模型输出的点评:
【GEMINI】该模型生成的《产品需求说明书》模板表现极其优秀。它不仅完整地履行了用户指令,采用了正确的工具(write)生成文件,而且在内容深度和广度上远超基础要求,提供了一个工业级标准的专业模板。文档结构严谨,排版美观,且具有极强的实操指导意义。
进阶难度评测结果
- 得分:100.0 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:
# 产品需求说明书 --- ## 封面页 - **文档名称**:产品需求说明书 - **项目名称**:[填写项目名称] - **版本号**:V1.0 - **编写人**:[填写] - **审核人**:[填写] - **日期**:[填写日期] --- ## 修订记录 | 版本号 | 修订日期 | 修订人 | 修订内容 | 审批人 | | ------ | -------- | ------ | -------- | ------ | | V1.0 | [日期] | [姓名] | 初始版本 | [姓名] | --- ## 目录 1. 引言 1.1 编写目的 1.2 术语与缩略语 2. 产品概述 2.1 产品背景 2.2 产品目标 2.3 用户画像 3. 功能需求 3.1 功能模块一 3.1.1 功能描述 3.1.2 输入/输出 3.1.3 业务规则 4. 非功能需求 4.1 性能需求 4.2 安全需求 4.3 可用性需求 5. 接口需求 5.1 外部接口 5.2 内部接口 6. 附录 --- ## 1. 引言 ### 1.1 编写目的 [说明本文档的编写目的及预期读者] ### 1.2 术语与缩略语 | 术语/缩略语 | 说明 | | ------------ | ---- | | [术语] | [说明] | --- ## 2. 产品概述 ### 2.1 产品背景 [描述产品的市场背景、业务驱动力及立项原因] ### 2.2 产品目标 [明确产品要达成的核心目标及衡量指标] ### 2.3 用户画像 [描述目标用户群体的特征、角色及核心诉求] --- ## 3. 功能需求 ### 3.1 功能模块一 #### 3.1.1 功能描述 [描述该功能的具体行为与交互流程] #### 3.1.2 输入/输出 | 项目 | 说明 | | ---- | ---- | | 输入 | [描述输入数据] | | 输出 | [描述输出结果] | #### 3.1.3 业务规则 [列出该功能涉及的业务规则与约束条件] --- ## 4. 非功能需求 ### 4.1 性能需求 [响应时间、吞吐量、并发量等要求] ### 4.2 安全需求 [数据安全、权限控制、合规性等要求] ### 4.3 可用性需求 [系统可用性指标、容灾与降级策略] --- ## 5. 接口需求 ### 5.1 外部接口 [与第三方系统或外部服务的接口说明] ### 5.2 内部接口 [系统内部各模块之间的接口说明] --- ## 6. 附录 [补充说明、参考资料、相关文档链接等] 文件已写入 `product_requirement_template.md`,三项调整均已落实: 1. **编号格式**:正文章节统一使用 `1` / `1.1` / `1.1.1` 层级编号,目录与正文保持一致。 2. **审批人列**:修订记录表格新增「审批人」列。 3. **加粗字段名**:封面页所有字段名(**文档名称**、**项目名称**、**版本号**、**编写人**、**审核人**、**日期**)均使用加粗格式。
AI 评审点评
以下是 AI 评审对该模型输出的点评:
【GEMINI】模型完美地完成了任务。它不仅准确执行了所有的格式调整指令,还通过工具调用生成了结构完整、专业度高的产品需求说明书模板,完全符合预期。
相关链接
您可以通过以下链接查看更多相关内容: