一键导入
hf-design-review
适用于设计草稿已完成需要正式 review verdict、或被指定为 reviewer subagent 执行设计评审的场景。不适用于需继续写或修设计(→ hf-design)、需拆任务或编码(→ hf-tasks)、阶段不清或证据冲突(→ hf-workflow-router)。
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
适用于设计草稿已完成需要正式 review verdict、或被指定为 reviewer subagent 执行设计评审的场景。不适用于需继续写或修设计(→ hf-design)、需拆任务或编码(→ hf-tasks)、阶段不清或证据冲突(→ hf-workflow-router)。
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Use when starting an existing-code bug audit on a repository or large directory tree. First detects project language + architecture (e.g. C/C++ embedded SOA, Python web service, frontend SPA), proposes a tailored review checklist (scenario-specific bug categories) for user confirmation, then slices the codebase into modules within a per-module token budget. Produces plan.json with profile + review_checklist + modules that downstream audit-reviewer consumes module-by-module. Not for PR diff review (use hf-code-review) or for actually finding bugs (use audit-reviewer).
Use as the FINAL stage of the code-audit pipeline. Reads confirmed.json (output of audit-verifier) and renders a self-contained single-file HTML report (always) and optionally an Excel workbook. The HTML contains summary stats by severity/category/module, filterable finding cards with code snippets and audit trails (reviewer + verifier). Not for emitting findings (use audit-reviewer) or verifying findings (use audit-verifier).
Use when scanning an existing-code module for bugs and emitting finding drafts. Reads source files within one module from the plan.json produced by audit-planner, walks files line-by-line, emits findings/<module>.json with file path, line numbers, category, severity, confidence, code snippet evidence, and reasoning. The set of allowed finding categories is sourced from plan.json's review_checklist (scenario preset such as c-cpp-embedded-soa / python-web-service / frontend-spa / generic) rather than a fixed taxonomy — keep findings scoped to the user-confirmed checklist. This is the PRIMARY (first-stage) reviewer in the two-agent confirmation pipeline; downstream audit-verifier independently confirms each finding. Not for PR diff review (use hf-code-review) or for verifying findings (use audit-verifier).
Use as the SECOND-STAGE independent confirmer in the two-agent code-audit pipeline. Reads finding drafts produced by audit-reviewer and independently re-examines each one against the actual source code, writes verifications/<module>.json with status (confirmed/rejected/upgrade/downgrade/needs_more_evidence), reason, and evidence_check. Operates with FRESH context — does not see reviewer's internal reasoning beyond what is recorded in the finding's description+evidence fields, to enable independent judgement. Not for emitting new findings (use audit-reviewer) or rendering the report (use audit-reporter).
Use when hf-test-driven-dev finishes GREEN on a frontend-touching active task whose spec declares a UI surface, and fresh browser runtime evidence (screenshot / console log / network trace) is required for downstream gates. Not for issuing verdicts (gates do that), not for replacing hf-test-review's test-quality review, not for backend-only tasks.
适用于 test review 通过后评审代码质量、用户要求 code review 的场景。不适用于评审测试(→ hf-test-review)、写/修代码(→ hf-test-driven-dev)、阶段不清(→ hf-workflow-router)。
| name | hf-design-review |
| description | 适用于设计草稿已完成需要正式 review verdict、或被指定为 reviewer subagent 执行设计评审的场景。不适用于需继续写或修设计(→ hf-design)、需拆任务或编码(→ hf-tasks)、阶段不清或证据冲突(→ hf-workflow-router)。 |
评审设计文档,判断它是否可以提交给 approval step。核心职责是防止过早拆解任务——确保设计真正锚定规格、决策站得住、接口边界清楚到足以进入任务规划。
职责边界:本 skill 只评审 架构 / 模块 / API 契约 / 数据模型 / 后端 NFR。若规格声明 UI surface 且 hf-ui-design 被激活,hf-ui-review 会作为 design stage 的并行 review peer 独立评审 IA / 交互 / 视觉 / 前端 a11y 等 UI 维度。两条 review 均通过后,父会话才发起联合 设计真人确认;任一未过,对应起草 skill 回修,另一条可继续其稳定部分。不得跨权评审 hf-ui-design 的职责范围;发现 peer 交接块不一致时,只在 findings 中标注,不代位评审。
本 skill 融合以下已验证方法:
使用:
hf-design 已完成设计草稿,需要正式 review verdict不使用:
hf-designhf-workflow-router直接调用信号:"review 这份设计"、"设计评审"、"帮我看一下这个设计"。
前提条件:存在当前设计草稿、已批准规格、项目相关约定(若项目已声明)。缺设计草稿 → hf-design;缺已批准规格或阶段不清 → hf-workflow-router。
读取:已批准规格(默认 features/<active>/spec.md)、被评审设计文档(默认 features/<active>/design.md)、项目级评审约定(若项目已声明)、feature progress.md(默认 features/<active>/progress.md)当前状态、最少必要技术上下文。
产出:review 记录正文 + 结构化 reviewer 返回摘要。
评审记录落盘由 reviewer 负责;approval step 和主链推进由父会话负责。
hf-tasks读取并固定证据来源:已批准规格(默认 features/<active>/spec.md)、当前设计文档(默认 features/<active>/design.md)、项目约定、feature progress.md(默认 features/<active>/progress.md)状态、必要技术上下文。
不要只根据会话记忆或零散聊天内容判断"已批准"或"设计已经讲清楚"。
检查:是否存在稳定可定位的设计草稿、已批准规格是否可回读、route/stage/profile 与 approval evidence 是否一致。
reroute_via_router=truehf-design对 6 个维度做内部评分(0-10):需求覆盖与追溯、架构一致性、决策质量、约束与 NFR 适配、接口与任务规划准备度、测试准备度与隐藏假设。
评分辅助区分:轻微缺口 vs 需修改 vs 阻塞。按 references/review-checklist.md 逐项审查。
每条 finding 必须带:
severity(critical / important / minor)classification(USER-INPUT / LLM-FIXABLE)rule_id(如 D3、D5、A2)默认分类:
USER-INPUT:缺失外部阈值、未确认业务裁决、规格未批准却引入关键新增能力、核心 trade-off 仍需真人拍板LLM-FIXABLE:缺少方案对比、接口边界说明不足、任务规划准备度表达不清、测试策略未显式写出、隐藏假设未整理成文判定规则(详见 references/review-record-template.md):
severity:critical(阻塞任务规划)> important(应修复)> minor(建议改进)。
按 references/review-record-template.md 写评审记录,并返回结构化 JSON 给父会话。
下一步映射:
通过 → 设计真人确认(needs_human_confirmation=true)需修改 → hf-design阻塞(设计内容) → hf-design阻塞(需求漂移/规格冲突) → hf-workflow-router(reroute_via_router=true)按需加载详细参考内容:
| 主题 | Reference | 加载时机 |
|---|---|---|
| 评审检查清单 | references/review-checklist.md | 执行 Step 2 多维审查时 |
| 评审记录模板 | references/review-record-template.md | 执行 Step 4 写评审记录时 |
hf-tasks(跳过 approval step)| 借口 | 反驳 / Hard rule |
|---|---|
| "Bounded Context 边界稍模糊但不影响实现。" | Hard Gates: DDD 战略边界含糊会在实现时反弹为模块边界争议;缺位 → finding。 |
| "ADR 我可以替作者补。" | Hard Gates (soul.md): 评审者不能补作者的工件;缺 ADR → finding,由作者回修。 |
| "STRIDE 没做但风险不大。" | Hard Gates: 缺 STRIDE → finding;不能由 reviewer 替评估为"风险不大"。 |
| "verdict pass,回头实现时再发现问题再说。" | Verification: design verdict 是后续 hf-tasks / hf-test-driven-dev 的 precondition;草率 pass 会回流大量返工。 |
完成条件:
next_action_or_recommended_skill通过,已明确要求进入 设计真人确认