In the OpenSpec project, parse user-provided design images, prototypes, screenshots, or architecture diagrams and convert visual information into structured descriptions; automatically load when requirement analysis involves the front end or contains images. Parsing results are incorporated into requirement descriptions and scenario descriptions in specs, and can also provide input for project-analysis and coding-implement.
本技能用于在 OpenSpec 体系下,对用户提供的图片(设计稿、原型图、界面截图、架构图、数据流图等)进行解析,将视觉与布局信息转化为可被需求、设计与实现环节使用的结构化文本描述。
补充方法论:解析完成后或当无图片需产出线框图、流程图、状态图、时序图等时,可使用 Mermaid 在 Markdown 中生成图表,存放于 design/documents/changes/[change-id]/design-assets/ 并在 PRD 中引用。书写规范、图表类型选择与自检见 REFERENCE《Mermaid 使用规范与技巧》:ai-agent-dev-system/skills/image-analysis/REFERENCE/mermaid-usage-norms-and-tips.md。
触发方式:
request-analysis 自动加载本技能;前置资料来源(解析前建议读取):
openspec/changes/[change-id]/ 下已有的 proposal.md、design.md、specs/*/spec.md(若已存在 change-id);design/documents/changes/[change-id]/ 下的功能需求说明书等(若已存在);openspec/project.md 与 design/project-rules/ 中与前端/UI/数据流相关的约定,便于解析结果与项目语境对齐。确认图片与上下文
change-id 的对应关系;按规范解析图片
image-analysis/REFERENCE/ 下的图片分析规范,对每张图片进行结构化解析;输出结构化描述
纳入需求与 spec
design/documents/changes/[change-id]/功能需求说明书.md(或等价文档)中的界面/交互描述;openspec/changes/[change-id]/specs/[capability]/spec.md 中相关 Requirement 的 Scenario 描述(布局、文案、状态、边界场景等);design/project-rules/ 或对应 change 的 design.md。design/documents/changes/[change-id]/design-assets/flows/ 或 wireframes/,并遵循 REFERENCE《Mermaid 使用规范与技巧》;在 PRD 或需求说明中引用对应 .md。标注来源与不确定性
project.md 或 design/project-rules/ 中架构与数据流描述的输入。design/documents/changes/[change-id]/ 与 openspec/changes/[change-id]/specs/*/spec.md 的场景描述。design/documents/changes/[change-id]/功能需求说明书.md 的「商品详情页」小节,并同步到 openspec/changes/[change-id]/specs/.../spec.md 中相关 Requirement 的 Scenario(如「当用户进入详情页时,页面呈现…」)。通过本技能,OpenSpec 项目可将「设计图/截图」与「需求说明、spec 场景」打通,减少信息丢失与理解偏差。
Search for places (restaurants, cafes, etc.) via Google Places API proxy on localhost.
Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.
Create or update AgentSkills. Use when designing, structuring, or packaging skills with scripts, references, and assets.
Start voice calls via the OpenClaw voice-call plugin.
Notion API for creating and managing pages, databases, and blocks.
Gemini CLI for one-shot Q&A, summaries, and generation.
Category:developer