# aipex - Doramagic AI Context Pack

> 定位：安装前体验与判断资产。它帮助宿主 AI 有一个好的开始，但不代表已经安装、执行或验证目标项目。

## 充分原则

- **充分原则，不是压缩原则**：AI Context Pack 应该充分到让宿主 AI 在开工前理解项目价值、能力边界、使用入口、风险和证据来源；它可以分层组织，但不以最短摘要为目标。
- **压缩策略**：只压缩噪声和重复内容，不压缩会影响判断和开工质量的上下文。

## 给宿主 AI 的使用方式

你正在读取 Doramagic 为 aipex 编译的 AI Context Pack。请把它当作开工前上下文：帮助用户理解适合谁、能做什么、如何开始、哪些必须安装后验证、风险在哪里。不要声称你已经安装、运行或执行了目标项目。

## Claim 消费规则

- **事实来源**：Repo Evidence + Claim/Evidence Graph；Human Wiki 只提供显著性、术语和叙事结构。
- **事实最低状态**：`supported`
- `supported`：可以作为项目事实使用，但回答中必须引用 claim_id 和证据路径。
- `weak`：只能作为低置信度线索，必须要求用户继续核实。
- `inferred`：只能用于风险提示或待确认问题，不能包装成项目事实。
- `unverified`：不得作为事实使用，应明确说证据不足。
- `contradicted`：必须展示冲突来源，不得替用户强行选择一个版本。

## 它最适合谁

- **正在使用 Claude/Codex/Cursor/Gemini 等宿主 AI 的开发者**：README 或插件配置提到多个宿主 AI。 证据：`README.md` Claim：`clm_0003` supported 0.86
- **希望把专业流程带进宿主 AI 的用户**：仓库包含 Skill 文档。 证据：`packages/browser-runtime/src/skill/built-in/skill-creator-browser/SKILL.md`, `packages/browser-runtime/src/skill/built-in/ux-audit-walkthrough/SKILL.md`, `packages/browser-runtime/src/skill/built-in/wcag22-a11y-audit/SKILL.md`, `skill/SKILL.md` Claim：`clm_0004` supported 0.86

## 它能做什么

- **AI Skill / Agent 指令资产库**（可做安装前预览）：项目包含可被宿主 AI 读取的 Skill 或 Agent 指令文件，可用于把专业流程带入 Claude、Codex、Cursor 等宿主。 证据：`packages/browser-runtime/src/skill/built-in/skill-creator-browser/SKILL.md`, `packages/browser-runtime/src/skill/built-in/ux-audit-walkthrough/SKILL.md`, `packages/browser-runtime/src/skill/built-in/wcag22-a11y-audit/SKILL.md`, `skill/SKILL.md` Claim：`clm_0001` supported 0.86
- **命令行启动或安装流程**（需要安装后验证）：项目文档中存在可执行命令，真实使用需要在本地或宿主环境中运行这些命令。 证据：`README.md`, `packages/aipex-react/README.md`, `packages/browser-runtime/README.md`, `packages/core/README.md` 等 Claim：`clm_0002` supported 0.86

## 怎么开始

- `claude mcp add aipex-browser -- npx -y aipex-mcp-bridge` 证据：`README.md` Claim：`clm_0005` supported 0.86
- `npm install @aipexstudio/aipex-react @aipexstudio/aipex-core` 证据：`packages/aipex-react/README.md` Claim：`clm_0006` supported 0.86
- `pnpm add @aipexstudio/aipex-react @aipexstudio/aipex-core` 证据：`packages/aipex-react/README.md` Claim：`clm_0007` supported 0.86
- `npm install @aipexstudio/browser-runtime` 证据：`packages/browser-runtime/README.md` Claim：`clm_0008` supported 0.86
- `pnpm add @aipexstudio/browser-runtime` 证据：`packages/browser-runtime/README.md` Claim：`clm_0009` supported 0.86
- `npm install @aipexstudio/aipex-core` 证据：`packages/core/README.md` Claim：`clm_0010` supported 0.86
- `pnpm add @aipexstudio/aipex-core` 证据：`packages/core/README.md` Claim：`clm_0011` supported 0.86
- `npm install @aipexstudio/dom-snapshot` 证据：`packages/dom-snapshot/README.md` Claim：`clm_0012` supported 0.86
- `pnpm add @aipexstudio/dom-snapshot` 证据：`packages/dom-snapshot/README.md` Claim：`clm_0013` supported 0.86

## 继续前判断卡

- **当前建议**：先做权限沙盒试用
- **为什么**：项目存在安装命令、宿主配置或本地写入线索，不建议直接进入主力环境，应先在隔离环境试装。

### 30 秒判断

- **现在怎么做**：先做权限沙盒试用
- **最小安全下一步**：先跑 Prompt Preview；若仍要安装，只在隔离环境试装
- **先别相信**：工具权限边界不能在安装前相信。
- **继续会触碰**：命令执行、宿主 AI 配置、本地环境或项目文件

### 现在可以相信

- **适合人群线索：正在使用 Claude/Codex/Cursor/Gemini 等宿主 AI 的开发者**（supported）：有 supported claim 或项目证据支撑，但仍不等于真实安装效果。 证据：`README.md` Claim：`clm_0003` supported 0.86
- **适合人群线索：希望把专业流程带进宿主 AI 的用户**（supported）：有 supported claim 或项目证据支撑，但仍不等于真实安装效果。 证据：`packages/browser-runtime/src/skill/built-in/skill-creator-browser/SKILL.md`, `packages/browser-runtime/src/skill/built-in/ux-audit-walkthrough/SKILL.md`, `packages/browser-runtime/src/skill/built-in/wcag22-a11y-audit/SKILL.md`, `skill/SKILL.md` Claim：`clm_0004` supported 0.86
- **能力存在：AI Skill / Agent 指令资产库**（supported）：可以相信项目包含这类能力线索；是否适合你的具体任务仍要试用或安装后验证。 证据：`packages/browser-runtime/src/skill/built-in/skill-creator-browser/SKILL.md`, `packages/browser-runtime/src/skill/built-in/ux-audit-walkthrough/SKILL.md`, `packages/browser-runtime/src/skill/built-in/wcag22-a11y-audit/SKILL.md`, `skill/SKILL.md` Claim：`clm_0001` supported 0.86
- **能力存在：命令行启动或安装流程**（supported）：可以相信项目包含这类能力线索；是否适合你的具体任务仍要试用或安装后验证。 证据：`README.md`, `packages/aipex-react/README.md`, `packages/browser-runtime/README.md`, `packages/core/README.md` 等 Claim：`clm_0002` supported 0.86
- **存在 Quick Start / 安装命令线索**（supported）：可以相信项目文档出现过启动或安装入口；不要因此直接在主力环境运行。 证据：`README.md` Claim：`clm_0005` supported 0.86

### 现在还不能相信

- **工具权限边界不能在安装前相信。**（unverified）：MCP/tool 类项目通常会触碰文件、网络、浏览器或外部 API，必须真实检查权限和日志。
- **真实输出质量不能在安装前相信。**（unverified）：Prompt Preview 只能展示引导方式，不能证明真实项目中的结果质量。
- **宿主 AI 版本兼容性不能在安装前相信。**（unverified）：Claude、Cursor、Codex、Gemini 等宿主加载规则和版本差异必须在真实环境验证。
- **不会污染现有宿主 AI 行为，不能直接相信。**（inferred）：Skill、plugin、AGENTS/CLAUDE/GEMINI 指令可能改变宿主 AI 的默认行为。 证据：`AGENTS.md`, `packages/browser-runtime/src/skill/built-in/skill-creator-browser/SKILL.md`, `packages/browser-runtime/src/skill/built-in/ux-audit-walkthrough/SKILL.md`, `packages/browser-runtime/src/skill/built-in/wcag22-a11y-audit/SKILL.md` 等
- **可安全回滚不能默认相信。**（unverified）：除非项目明确提供卸载和恢复说明，否则必须先在隔离环境验证。
- **真实安装后是否与用户当前宿主 AI 版本兼容？**（unverified）：兼容性只能通过实际宿主环境验证。
- **项目输出质量是否满足用户具体任务？**（unverified）：安装前预览只能展示流程和边界，不能替代真实评测。
- **安装命令是否需要网络、权限或全局写入？**（unverified）：这影响企业环境和个人环境的安装风险。 证据：`README.md`

### 继续会触碰什么

- **命令执行**：包管理器、网络下载、本地插件目录、项目配置或用户主目录。 原因：运行第一条命令就可能产生环境改动；必须先判断是否值得跑。 证据：`README.md`, `packages/aipex-react/README.md`, `packages/browser-runtime/README.md`, `packages/core/README.md` 等
- **宿主 AI 配置**：Claude/Codex/Cursor/Gemini/OpenCode 等宿主的 plugin、Skill 或规则加载配置。 原因：宿主配置会改变 AI 后续工作方式，可能和用户已有规则冲突。 证据：`AGENTS.md`, `packages/browser-runtime/src/skill/built-in/skill-creator-browser/SKILL.md`, `packages/browser-runtime/src/skill/built-in/ux-audit-walkthrough/SKILL.md`, `packages/browser-runtime/src/skill/built-in/wcag22-a11y-audit/SKILL.md` 等
- **本地环境或项目文件**：安装结果、插件缓存、项目配置或本地依赖目录。 原因：安装前无法证明写入范围和回滚方式，需要隔离验证。 证据：`README.md`, `packages/aipex-react/README.md`, `packages/browser-runtime/README.md`, `packages/core/README.md` 等
- **宿主 AI 上下文**：AI Context Pack、Prompt Preview、Skill 路由、风险规则和项目事实。 原因：导入上下文会影响宿主 AI 后续判断，必须避免把未验证项包装成事实。

### 最小安全下一步

- **先跑 Prompt Preview**：用安装前交互式试用判断工作方式是否匹配，不需要授权或改环境。（适用：任何项目都适用，尤其是输出质量未知时。）
- **只在隔离目录或测试账号试装**：避免安装命令污染主力宿主 AI、真实项目或用户主目录。（适用：存在命令执行、插件配置或本地写入线索时。）
- **先备份宿主 AI 配置**：Skill、plugin、规则文件可能改变 Claude/Cursor/Codex 的默认行为。（适用：存在插件 manifest、Skill 或宿主规则入口时。）
- **安装后只验证一个最小任务**：先验证加载、兼容、输出质量和回滚，再决定是否深用。（适用：准备从试用进入真实工作流时。）

### 退出方式

- **保留安装前状态**：记录原始宿主配置和项目状态，后续才能判断是否可恢复。
- **准备移除宿主 plugin / Skill / 规则入口**：如果试装后行为异常，可以把宿主 AI 恢复到试装前状态。
- **记录安装命令和写入路径**：没有明确卸载说明时，至少要知道哪些目录或配置需要手动清理。
- **如果没有回滚路径，不进入主力环境**：不可回滚是继续前阻断项，不应靠信任或运气继续。

## 哪些只能预览

- 解释项目适合谁和能做什么
- 基于项目文档演示典型对话流程
- 帮助用户判断是否值得安装或继续研究

## 哪些必须安装后验证

- 真实安装 Skill、插件或 CLI
- 执行脚本、修改本地文件或访问外部服务
- 验证真实输出质量、性能和兼容性

## 边界与风险判断卡

- **把安装前预览误认为真实运行**：用户可能高估项目已经完成的配置、权限和兼容性验证。 处理方式：明确区分 prompt_preview_can_do 与 runtime_required。 Claim：`clm_0014` inferred 0.45
- **命令执行会修改本地环境**：安装命令可能写入用户主目录、宿主插件目录或项目配置。 处理方式：先在隔离环境或测试账号中运行。 证据：`README.md`, `packages/aipex-react/README.md`, `packages/browser-runtime/README.md`, `packages/core/README.md` 等 Claim：`clm_0015` supported 0.86
- **待确认**：真实安装后是否与用户当前宿主 AI 版本兼容？。原因：兼容性只能通过实际宿主环境验证。
- **待确认**：项目输出质量是否满足用户具体任务？。原因：安装前预览只能展示流程和边界，不能替代真实评测。
- **待确认**：安装命令是否需要网络、权限或全局写入？。原因：这影响企业环境和个人环境的安装风险。

## 开工前工作上下文

### 加载顺序

- 先读取 how_to_use.host_ai_instruction，建立安装前判断资产的边界。
- 读取 claim_graph_summary，确认事实来自 Claim/Evidence Graph，而不是 Human Wiki 叙事。
- 再读取 intended_users、capabilities 和 quick_start_candidates，判断用户是否匹配。
- 需要执行具体任务时，优先查 role_skill_index，再查 evidence_index。
- 遇到真实安装、文件修改、网络访问、性能或兼容性问题时，转入 risk_card 和 boundaries.runtime_required。

### 任务路由

- **AI Skill / Agent 指令资产库**：先基于 role_skill_index / evidence_index 帮用户挑选可用角色、Skill 或工作流。 边界：可做安装前 Prompt 体验。 证据：`packages/browser-runtime/src/skill/built-in/skill-creator-browser/SKILL.md`, `packages/browser-runtime/src/skill/built-in/ux-audit-walkthrough/SKILL.md`, `packages/browser-runtime/src/skill/built-in/wcag22-a11y-audit/SKILL.md`, `skill/SKILL.md` Claim：`clm_0001` supported 0.86
- **命令行启动或安装流程**：先说明这是安装后验证能力，再给出安装前检查清单。 边界：必须真实安装或运行后验证。 证据：`README.md`, `packages/aipex-react/README.md`, `packages/browser-runtime/README.md`, `packages/core/README.md` 等 Claim：`clm_0002` supported 0.86

### 上下文规模

- 文件总数：485
- 重要文件覆盖：40/485
- 证据索引条目：75
- 角色 / Skill 条目：4

### 证据不足时的处理

- **missing_evidence**：说明证据不足，要求用户提供目标文件、README 段落或安装后验证记录；不要补全事实。
- **out_of_scope_request**：说明该任务超出当前 AI Context Pack 证据范围，并建议用户先查看 Human Manual 或真实安装后验证。
- **runtime_request**：给出安装前检查清单和命令来源，但不要替用户执行命令或声称已执行。
- **source_conflict**：同时展示冲突来源，标记为待核实，不要强行选择一个版本。

## Prompt Recipes

### 适配判断

- 目标：判断这个项目是否适合用户当前任务。
- 预期输出：适配结论、关键理由、证据引用、安装前可预览内容、必须安装后验证内容、下一步建议。

```text
请基于 aipex 的 AI Context Pack，先问我 3 个必要问题，然后判断它是否适合我的任务。回答必须包含：适合谁、能做什么、不能做什么、是否值得安装、证据来自哪里。所有项目事实必须引用 evidence_refs、source_paths 或 claim_id。
```

### 安装前体验

- 目标：让用户在安装前感受核心工作流，同时避免把预览包装成真实能力或营销承诺。
- 预期输出：一段带边界标签的体验剧本、安装后验证清单和谨慎建议；不含真实运行承诺或强营销表述。

```text
请把 aipex 当作安装前体验资产，而不是已安装工具或真实运行环境。

请严格输出四段：
1. 先问我 3 个必要问题。
2. 给出一段“体验剧本”：用 [安装前可预览]、[必须安装后验证]、[证据不足] 三种标签展示它可能如何引导工作流。
3. 给出安装后验证清单：列出哪些能力只有真实安装、真实宿主加载、真实项目运行后才能确认。
4. 给出谨慎建议：只能说“值得继续研究/试装”“先补充信息后再判断”或“不建议继续”，不得替项目背书。

硬性边界：
- 不要声称已经安装、运行、执行测试、修改文件或产生真实结果。
- 不要写“自动适配”“确保通过”“完美适配”“强烈建议安装”等承诺性表达。
- 如果描述安装后的工作方式，必须使用“如果安装成功且宿主正确加载 Skill，它可能会……”这种条件句。
- 体验剧本只能写成“示例台词/假设流程”：使用“可能会询问/可能会建议/可能会展示”，不要写“已写入、已生成、已通过、正在运行、正在生成”。
- Prompt Preview 不负责给安装命令；如用户准备试装，只能提示先阅读 Quick Start 和 Risk Card，并在隔离环境验证。
- 所有项目事实必须来自 supported claim、evidence_refs 或 source_paths；inferred/unverified 只能作风险或待确认项。

```

### 角色 / Skill 选择

- 目标：从项目里的角色或 Skill 中挑选最匹配的资产。
- 预期输出：候选角色或 Skill 列表，每项包含适用场景、证据路径、风险边界和是否需要安装后验证。

```text
请读取 role_skill_index，根据我的目标任务推荐 3-5 个最相关的角色或 Skill。每个推荐都要说明适用场景、可能输出、风险边界和 evidence_refs。
```

### 风险预检

- 目标：安装或引入前识别环境、权限、规则冲突和质量风险。
- 预期输出：环境、权限、依赖、许可、宿主冲突、质量风险和未知项的检查清单。

```text
请基于 risk_card、boundaries 和 quick_start_candidates，给我一份安装前风险预检清单。不要替我执行命令，只说明我应该检查什么、为什么检查、失败会有什么影响。
```

### 宿主 AI 开工指令

- 目标：把项目上下文转成一次对话开始前的宿主 AI 指令。
- 预期输出：一段边界明确、证据引用明确、适合复制给宿主 AI 的开工前指令。

```text
请基于 aipex 的 AI Context Pack，生成一段我可以粘贴给宿主 AI 的开工前指令。这段指令必须遵守 not_runtime=true，不能声称项目已经安装、运行或产生真实结果。
```

## 角色 / Skill 索引

- 共索引 4 个角色 / Skill / 项目文档条目。

- **skill-creator**（skill）：Guide for creating effective skills. This skill should be used when users want to create a new skill or update an existing skill that extends AIPex's capabilities with specialized knowledge, workflows, or tool integrations. 激活提示：当用户任务与“skill-creator”描述的流程高度相关时，先用它做安装前体验，再决定是否安装。 证据：`packages/browser-runtime/src/skill/built-in/skill-creator-browser/SKILL.md`
- **ux-audit-walkthrough**（skill）：Minimalist UX/Interaction Audit Expert that deconstructs complex interactions through cognitive load and operational efficiency lenses. Use this skill when you need to perform a UX walkthrough audit on a Figma prototype or web interface, evaluating usability based on principles like fewer clicks, less UI elements, no hidden logic, and self-explanatory design. 激活提示：当用户任务与“ux-audit-walkthrough”描述的流程高度相关时，先用它做安装前体验，再决定是否安装。 证据：`packages/browser-runtime/src/skill/built-in/ux-audit-walkthrough/SKILL.md`
- **wcag22-a11y-audit**（skill）：WCAG 2.2 Accessibility Audit skill that systematically evaluates web pages against 8 core Success Criteria 1.1.1, 1.4.3, 1.4.11, 2.1.1, 2.1.2, 2.4.3, 2.4.7, 4.1.2 using accessibility tree inspection and visual analysis. Use this skill when you need to perform accessibility testing/auditing on a live webpage. 激活提示：当用户任务与“wcag22-a11y-audit”描述的流程高度相关时，先用它做安装前体验，再决定是否安装。 证据：`packages/browser-runtime/src/skill/built-in/wcag22-a11y-audit/SKILL.md`
- **aipex-browser**（skill）：AI-powered browser automation using the AIPex Chrome Extension via MCP bridge. Use this skill when the agent needs to control a Chrome browser — navigating pages, clicking elements, filling forms, capturing screenshots, managing tabs, or downloading content — by connecting to the AIPex MCP bridge. 激活提示：当用户任务与“aipex-browser”描述的流程高度相关时，先用它做安装前体验，再决定是否安装。 证据：`skill/SKILL.md`

## 证据索引

- 共索引 75 条证据。

- **Why We Built This**（documentation）：English 中文 Deutsch Español Français 日本語 한국어 Português Русский 证据：`README.md`
- **aipex-mcp-bridge**（documentation）：MCP server that connects AI agents to the AIPex https://aipex.ai browser extension. Supports multiple simultaneous clients Cursor, Claude Code, VS Code Copilot, etc. via StreamableHTTP. 证据：`mcp-bridge/README.md`
- **@aipexstudio/aipex-react**（documentation）：React UI toolkit for building AIPex-powered chat and extension experiences. 证据：`packages/aipex-react/README.md`
- **AIPex Browser Extension @aipexstudio/cool-aipex**（documentation）：AIPex Browser Extension @aipexstudio/cool-aipex 证据：`packages/browser-ext/README.md`
- **@aipexstudio/browser-runtime**（documentation）：Chrome/Chromium runtime implementations for @aipexstudio/aipex-core . 证据：`packages/browser-runtime/README.md`
- **Conversation Storage**（documentation）：A robust conversation storage system for managing chat history with IndexedDB, LRU policy, and automatic migration from localStorage. 证据：`packages/browser-runtime/src/conversation/README.md`
- **@aipexstudio/aipex-core**（documentation）：Platform-agnostic TypeScript building blocks for creating streaming, tool-using AI agents . 证据：`packages/core/README.md`
- **@aipexstudio/dom-snapshot**（documentation）：A lightweight library for capturing DOM snapshots without relying on Chrome DevTools Protocol CDP Accessibility Tree AXTree . This library provides a pure JavaScript/TypeScript solution for creating structured page snapshots that can be used for web automation, testing, and AI-powered browser agents. 证据：`packages/dom-snapshot/README.md`
- **Package**（package_manifest）：{ "name": "aipex-mcp-bridge", "version": "3.1.0", "description": "MCP bridge that connects AI agents Cursor, Claude Code, VS Code Copilot, etc. to the AIPex browser extension. Auto-spawns a shared daemon to support multiple simultaneous clients.", "type": "module", "bin": { "aipex-mcp-bridge": "./dist/bridge.js", "aipex-mcp-daemon": "./dist/daemon.js", "aipex-cli": "./dist/cli.js" }, "files": "dist", "README.md" , "scripts": { "build": "tsup", "dev": "tsx src/bridge.ts", "dev:daemon": "tsx src/daemon.ts" }, "dependencies": { "@modelcontextprotocol/sdk": "^1.28.0", "ws": "^8.18.0" }, "devDependencies": { "@types/ws": "^8.5.13", "@types/node": "^22.0.0", "tsup": "^8.0.0", "tsx": "^4.21.0", "t… 证据：`mcp-bridge/package.json`
- **Package**（package_manifest）：{ "name": "@aipexstudio/root", "displayName": "AIPex", "version": "0.0.2", "private": true, "description": "Automate your browser with natural language commands - The open source browser-use solution", "author": "AIPex Studio", "repository": { "type": "git", "url": "git+https://github.com/AIPexStudio/AIPex.git" }, "scripts": { "postinstall": "prek install", "build": "pnpm -r --if-present build", "dev": "pnpm --filter @aipexstudio/cool-aipex dev", "preflight": "npm run format && npm run lint:fix && npm run typecheck && npm run test", "format": "biome format . --write", "format:check": "biome format .", "lint": "biome check .", "lint:fix": "biome check . --fix --unsafe", "lint:dependencies":… 证据：`package.json`
- **Package**（package_manifest）：{ "name": "@aipexstudio/aipex-react", "version": "0.1.0", "description": "React UI toolkit, hooks, and adapters for building AIPex-powered chat experiences", "main": "./dist/src/index.js", "types": "./dist/src/index.d.ts", "repository": { "type": "git", "url": "git+https://github.com/AIPexStudio/AIPex.git" }, "exports": { ".": { "types": "./dist/src/index.d.ts", "import": "./dist/src/index.js" }, "./adapters/ ": { "types": "./dist/src/adapters/ .d.ts", "import": "./dist/src/adapters/ .js" }, "./hooks/ ": { "types": "./dist/src/hooks/ .d.ts", "import": "./dist/src/hooks/ .js" }, "./components/chatbot": { "types": "./dist/src/components/chatbot/index.d.ts", "import": "./dist/src/components/ch… 证据：`packages/aipex-react/package.json`
- **Package**（package_manifest）：{ "name": "@aipexstudio/cool-aipex", "displayName": "AIPex", "version": "0.1.0", "description": "Automate your browser with natural language commands - The open source browser-use solution", "author": "AIPex Studio", "private": true, "repository": { "type": "git", "url": "git+https://github.com/AIPexStudio/AIPex.git" }, "workspaces": "packages/ " , "scripts": { "dev": "vite", "build": "vite build", "preview": "vite preview", "build:css": "npx tailwindcss -i ./src/tailwind.css -o ./src/style.css --minify", "typecheck": "tsc --project tsconfig.json", "test": "vitest run", "test:watch": "vitest" }, "files": "build/", "README.md", "package.json", "host-access-config.json" , "dependencies": { "@… 证据：`packages/browser-ext/package.json`
- **Package**（package_manifest）：{ "name": "@aipexstudio/browser-runtime", "version": "0.1.0", "description": "Browser automation runtime, context providers, and extension host contracts for AIPex", "main": "./dist/src/index.js", "types": "./dist/src/index.d.ts", "repository": { "type": "git", "url": "git+https://github.com/AIPexStudio/AIPex.git" }, "exports": { ".": { "types": "./dist/src/index.d.ts", "import": "./dist/src/index.js" }, "./hooks": { "types": "./dist/src/hooks/index.d.ts", "import": "./dist/src/hooks/index.js" } }, "files": "dist/src/ / .js", "dist/src/ / .d.ts", "!dist/src/ / .test.js", "!dist/src/ / .test.d.ts" , "author": "AIPex Studio", "scripts": { "build": "tsc -b", "typecheck": "tsc --project tsconfi… 证据：`packages/browser-runtime/package.json`
- **Package**（package_manifest）：{ "name": "@aipexstudio/aipex-core", "version": "0.1.0", "description": "AIPex Agent - Pure TypeScript AI Agent Framework", "main": "./dist/src/index.js", "types": "./dist/src/index.d.ts", "repository": { "type": "git", "url": "git+https://github.com/AIPexStudio/AIPex.git" }, "exports": { ".": { "types": "./dist/src/index.d.ts", "import": "./dist/src/index.js" } }, "files": "dist/src/ / .js", "dist/src/ / .d.ts", "!dist/src/ / .test.js", "!dist/src/ / .test.d.ts" , "scripts": { "build": "tsc", "test": "vitest run", "typecheck": "tsc --project tsconfig.json", "prepublishOnly": "npm run build", "example:basic": "tsx examples/basic-example.ts", "example:fork": "tsx examples/fork-example.ts", "… 证据：`packages/core/package.json`
- **Package**（package_manifest）：{ "name": "@aipexstudio/dom-snapshot", "version": "0.1.0", "description": "DOM snapshot utility for capturing and serializing web page state", "main": "./dist/src/index.js", "types": "./dist/src/index.d.ts", "repository": { "type": "git", "url": "git+https://github.com/AIPexStudio/AIPex.git" }, "exports": { ".": { "types": "./dist/src/index.d.ts", "import": "./dist/src/index.js" } }, "files": "dist/src/ / .js", "dist/src/ / .d.ts", "!dist/src/ / .test.js", "!dist/src/ / .test.d.ts" , "scripts": { "build": "tsc", "test": "vitest run", "typecheck": "tsc --project tsconfig.json", "prepublishOnly": "npm run build" }, "keywords": "dom", "snapshot", "serialize", "web", "browser" , "author": "AIPe… 证据：`packages/dom-snapshot/package.json`
- **Package Architecture**（documentation）：Key rule : @aipex-react must NOT depend on @browser-runtime . Browser-specific code ChromeStorageAdapter, browser tools stays in @browser-runtime or browser-ext . 证据：`AGENTS.md`
- **AIPex Browser Control**（skill_instruction）：AIPex is a Chrome extension that exposes 30+ browser automation tools over the Model Context Protocol MCP . Once connected, the agent can control any Chrome tab using natural language — clicking, typing, navigating, capturing screenshots, downloading content, and more. 证据：`skill/SKILL.md`
- **Skill Creator**（skill_instruction）：This skill provides guidance for creating effective skills. 证据：`packages/browser-runtime/src/skill/built-in/skill-creator-browser/SKILL.md`
- **UX Audit Walkthrough Skill**（skill_instruction）：Use this skill when the user wants to: 证据：`packages/browser-runtime/src/skill/built-in/ux-audit-walkthrough/SKILL.md`
- **WCAG 2.2 Accessibility Audit Skill**（skill_instruction）：Use this skill when the user wants to: 证据：`packages/browser-runtime/src/skill/built-in/wcag22-a11y-audit/SKILL.md`
- **License**（source_file）：Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files the "Software" , to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: 证据：`LICENSE`
- **Utils**（source_file）：export function formatBytes bytes: number : string export function formatDate date: Date : string export function getFileExtension filename: string : string / Get file icon based on extension / export function getFileIcon filename: string : string ⋮---- // Documents ⋮---- // Code ⋮---- // Images ⋮---- // Archives ⋮---- // Other ⋮---- / Check if a path is protected should not be deleted / export function isProtectedPath path: string : boolean 证据：`packages/aipex-react/src/components/file-manager/utils.ts`
- **Screenshot Utils**（source_file）：export function isValidScreenshotUid uid: string : boolean export function isCaptureScreenshotTool toolName: string : boolean export interface ScreenshotExtraction { imageData: string null; sendToLLM: boolean; screenshotUid: string null; } export function extractScreenshotFromToolResult toolName: string, result: unknown, : ScreenshotExtraction null function extractFromStructuredArray arr: unknown , : ScreenshotExtraction null export function transformScreenshotPlaceholders text: string, screenshotUidList: string , : string 证据：`packages/aipex-react/src/lib/screenshot-utils.ts`
- **Utils**（source_file）：import { type ClassValue, clsx } from "clsx"; import { twMerge } from "tailwind-merge"; export function cn ...inputs: ClassValue 证据：`packages/aipex-react/src/lib/utils.ts`
- **Automation Mode Toolbar**（source_file）：import { type AutomationMode, STORAGE KEYS, validateAutomationMode, } from "@aipexstudio/aipex-core"; import type { InputToolbarSlotProps } from "@aipexstudio/aipex-react"; import { TokenUsageIndicator } from "@aipexstudio/aipex-react/components/chatbot"; import { Button } from "@aipexstudio/aipex-react/components/ui/button"; import { DropdownMenu, DropdownMenuContent, DropdownMenuItem, DropdownMenuTrigger, } from "@aipexstudio/aipex-react/components/ui/dropdown-menu"; import { Tooltip, TooltipContent, TooltipTrigger, } from "@aipexstudio/aipex-react/components/ui/tooltip"; import { useTranslation } from "@aipexstudio/aipex-react/i18n/context"; import { cn } from "@aipexstudio/aipex-react/l… 证据：`packages/browser-ext/src/lib/automation-mode-toolbar.tsx`
- **Message Adapter**（source_file）：import type { UIMessage as ReactUIMessage } from "@aipexstudio/aipex-react/types"; import type { UIMessage as RuntimeUIMessage } from "@aipexstudio/browser-runtime"; ⋮---- interface ScreenshotToolInfo { imageData: string null; screenshotUid: string null; } function getScreenshotActualData parsedOutput: unknown, : Record null function extractScreenshotInfo toolName: string, parsedOutput: unknown, : ScreenshotToolInfo null function stripImageDataFromToolOutput toolName: string, content: string, : string export function toStorageFormat messages: ReactUIMessage , : RuntimeUIMessage function safeJsonParse value: unknown : T undefined function extractBusinessFailure result: unknown, : export func… 证据：`packages/browser-ext/src/lib/message-adapter.ts`
- **Filepreview**（source_file）：import { Alert, AlertDescription, } from "@aipexstudio/aipex-react/components/ui/alert"; import { Badge } from "@aipexstudio/aipex-react/components/ui/badge"; import { Button } from "@aipexstudio/aipex-react/components/ui/button"; import { Dialog, DialogContent, DialogDescription, DialogHeader, DialogTitle, } from "@aipexstudio/aipex-react/components/ui/dialog"; import { Textarea } from "@aipexstudio/aipex-react/components/ui/textarea"; import { useTheme } from "@aipexstudio/aipex-react/theme/context"; import type { FileInfo } from "@aipexstudio/browser-runtime"; import { skillManager, zenfs } from "@aipexstudio/browser-runtime"; import { AlertCircle, Code, Edit, File as FileIcon, FileText,… 证据：`packages/browser-ext/src/pages/options/file-components/FilePreview.tsx`
- **Utils**（source_file）：export function formatBytes bytes: number : string export function formatDate date: Date : string export function getFileExtension filename: string : string / Get file icon based on extension / export function getFileIcon filename: string : string ⋮---- // Documents ⋮---- // Code ⋮---- // Images ⋮---- // Archives ⋮---- // Other ⋮---- / Check if a path is protected should not be deleted / export function isProtectedPath path: string : boolean 证据：`packages/browser-ext/src/pages/options/file-components/utils.ts`
- **Mcp Bridge Panel**（source_file）：import type { WsMcpServerState } from "@aipexstudio/browser-runtime"; import { useCallback, useEffect, useState } from "react"; ⋮---- type ConnectionStatus = WsMcpServerState "status" ; ⋮---- const handleConnect = async = const handleDisconnect = async = ⋮---- onChange= 证据：`packages/browser-ext/src/pages/options/mcp-bridge-panel.tsx`
- **Share Conversation**（source_file）：import type { UIMessage } from "@aipexstudio/aipex-react/types"; import { WEBSITE URL } from "../config/website"; import { getAuthCookieHeader } from "./web-auth"; interface ShareablePart { type: string; key: string : unknown; } interface ShareableMessage { id: string; role: string; content: { parts: ShareablePart }; metadata?: Record ; } export interface ShareResult { url: string; } function toShareablePart part: UIMessage "parts" number : ShareablePart export async function shareConversation messages: UIMessage , : Promise 证据：`packages/browser-ext/src/services/share-conversation.ts`
- **Tool Manager**（source_file）：import type { FunctionTool } from "@aipexstudio/aipex-core"; import { allBrowserTools } from "@aipexstudio/browser-runtime"; export interface ToolMetadata { name: string; description: string; category: string; inputSchema: unknown; examples?: string ; } export interface AITool { type: "function"; function: { name: string; description: string; parameters: unknown; }; } export type ToolEventType = "tool registered" "tool unregistered"; type ToolSubscriber = data: unknown = void; ⋮---- export type ToolCategoryType = typeof ToolCategory keyof typeof ToolCategory ; export class ToolManager ⋮---- private constructor public static getInstance : ToolManager public getAllTools : FunctionTool public… 证据：`packages/browser-ext/src/services/tool-manager.ts`
- **Puppeteer Test Utils**（source_file）：import type { Browser, Page } from "puppeteer"; import puppeteer from "puppeteer"; interface ChromeDebuggerShim { sendCommand: target: { tabId: number }, method: string, params?: Record , callback?: result: unknown = void, = void; attach: target: { tabId: number }, requiredVersion: string, callback?: = void, = void; detach: target: { tabId: number }, callback?: = void = void; onDetach: { addListener: callback: source: { tabId?: number }, reason: string = void, = void; }; } interface TestContext { browser: Browser; page: Page; tabId: number; chrome: { debugger: ChromeDebuggerShim; runtime: { lastError: { message: string } undefined; }; }; cleanup: = Promise ; } ⋮---- async function createChr… 证据：`packages/browser-runtime/src/automation/__tests__/puppeteer-test-utils.ts`
- **Zenfs Manager**（source_file）：import { configure, fs } from "@zenfs/core"; import { IndexedDB } from "@zenfs/dom"; import type { FileStats } from "./types"; type FsPromises = { mkdir: ...args: any = Promise ; readFile: ...args: any = Promise ; writeFile: ...args: any = Promise ; stat: ...args: any = Promise ; readdir: ...args: any = Promise ; rmdir: ...args: any = Promise ; unlink: ...args: any = Promise ; }; class ZenFSManager ⋮---- async initialize : Promise private async initialize : Promise private getFsPromises : FsPromises private async ensureInitialized : Promise getSkillPath skillId: string : string async readFile path: string, encoding?: BufferEncoding, : Promise async writeFile path: string, data: string Uint8… 证据：`packages/browser-runtime/src/lib/vm/zenfs-manager.ts`
- **Skill Manager**（source_file）：import { zenfs } from "../../../lib/vm/zenfs-manager"; import licenseText from "../../built-in/skill-creator-browser/LICENSE.txt?raw"; import skillCreatorMarkdown from "../../built-in/skill-creator-browser/SKILL.md?raw"; import deleteFileScript from "../../built-in/skill-creator-browser/scripts/delete file.js?raw"; import initSkillScript from "../../built-in/skill-creator-browser/scripts/init skill.js?raw"; import packageSkillScript from "../../built-in/skill-creator-browser/scripts/package skill.js?raw"; import validateSkillScript from "../../built-in/skill-creator-browser/scripts/quick validate.js?raw"; import writeFileScript from "../../built-in/skill-creator-browser/scripts/write file.j… 证据：`packages/browser-runtime/src/skill/lib/services/skill-manager.ts`
- **Zip Utils**（source_file）：import { strFromU8, unzipSync } from "fflate"; import { zenfs } from "../../../lib/vm/zenfs-manager"; export interface ParsedSkillMetadata { name: string; description: string; version: string; } export class SkillConflictError extends Error ⋮---- constructor skillName: string ⋮---- export function parseSkillMetadata markdown: string : ParsedSkillMetadata export async function parseSkillMetadataFromZip zipBlob: Blob, : Promise function shouldFilterPath path: string : boolean function detectTopLevelDirectory paths: string : string ⋮---- // Check if all paths start with the same directory ⋮---- // Check if all paths start with this directory ⋮---- / Extract ZIP file directly to ZenFS at the sp… 证据：`packages/browser-runtime/src/skill/lib/utils/zip-utils.ts`
- **Tab Utils**（source_file）：export async function getActiveTab : Promise export async function executeScriptInTab tabId: number, func: ...args: Args = T, args: Args, : Promise export async function executeScriptInActiveTab func: ...args: Args = T, args: Args, : Promise 证据：`packages/browser-runtime/src/tools/tab-utils.ts`
- **Utils**（source_file）：import type { ConversationConfig } from "../types.js"; import { DEFAULT CONVERSATION CONFIG } from "./defaults.js"; ⋮---- export function createConversationConfig overrides?: ConversationConfig, : ConversationConfig export function normalizeConversationConfig value: unknown, : ConversationConfig export function isValidConversationStorage value: unknown, : value is ConversationConfig "storage" 证据：`packages/core/src/config/utils.ts`
- **Utils**（source_file）：import type { Context } from "./types"; export function formatContextsForPrompt contexts: Context : string ⋮---- // Convert value to string if it's not already ⋮---- export function isContext value: unknown : value is Context export async function resolveContexts contextOrIds: Context string , getContext: id: string = Promise , : Promise 证据：`packages/core/src/context/utils.ts`
- **Warum wir das gebaut haben**（documentation）：English 中文 Deutsch Español Français 日本語 한국어 Português Русский 证据：`README.de.md`
- **Por qué construimos esto**（documentation）：English 中文 Deutsch Español Français 日本語 한국어 Português Русский 证据：`README.es.md`
- **Pourquoi nous avons construit ceci**（documentation）：English 中文 Deutsch Español Français 日本語 한국어 Português Русский 证据：`README.fr.md`
- **これを作った理由**（documentation）：English 中文 Deutsch Español Français 日本語 한국어 Português Русский 证据：`README.ja.md`
- **이것을 만든 이유**（documentation）：English 中文 Deutsch Español Français 日本語 한국어 Português Русский 证据：`README.ko.md`
- **Por que construímos isso**（documentation）：English 中文 Deutsch Español Français 日本語 한국어 Português Русский 证据：`README.pt.md`
- **Почему мы создали это**（documentation）：English 中文 Deutsch Español Français 日本語 한국어 Português Русский 证据：`README.ru.md`
- **AIPex**（documentation）：你的下一个AI浏览器，何必需要迁移？ 开源、免费、隐私保护。直接在你现有的浏览器里跑。 证据：`README.zh-CN.md`
- **AIPex vs new-aipex: Logic Inconsistencies by Package**（documentation）：AIPex vs new-aipex: Logic Inconsistencies by Package 证据：`migration/LOGIC_INCONSISTENCIES_BY_PACKAGE.md`
- **AIPex Private 分支功能迁移策略**（documentation）：版本 : 2.0 创建日期 : 2026-01-03 源分支 : remotes/private/private 目标分支 : feature-next-rob 架构模式 : 多包架构 Monorepo 证据：`migration/MIGRATION_STRATEGY.md`
- **Sidepanel 入口链路：new-aipex vs aipex 缺口分析**（documentation）：Sidepanel 入口链路：new-aipex vs aipex 缺口分析 证据：`migration/SIDEPANEL_ENTRY_GAP_ANALYSIS.md`
- **Tool Surface Audit: aipex vs new-aipex**（documentation）：Tool Surface Audit: aipex vs new-aipex 证据：`migration/TOOL_SURFACE_AUDIT.md`
- **Compatibility Verification: aipex vs new-aipex Conversation Storage**（documentation）：Compatibility Verification: aipex vs new-aipex Conversation Storage 证据：`packages/browser-runtime/src/conversation/COMPATIBILITY.md`
- **AIPex Browser Tools Reference**（documentation）：Complete parameter schemas and usage examples for all AIPex MCP tools exposed via aipex-mcp-bridge . 证据：`skill/references/tools-reference.md`
- **Biome**（structured_config）：{ "root": true, "$schema": "https://biomejs.dev/schemas/2.3.14/schema.json", "formatter": { "enabled": true, "indentStyle": "space", "indentWidth": 2 }, "assist": { "actions": { "source": { "organizeImports": "on" } } }, "css": { "parser": { "tailwindDirectives": true } }, "linter": { "enabled": true, "rules": { "recommended": true, "a11y": { "useSemanticElements": "off" }, "suspicious": { "noExplicitAny": "off" }, "style": { "noNonNullAssertion": "off" }, "complexity": { "useLiteralKeys": "off" } } }, "overrides": { "includes": " / .json" , "formatter": { "indentStyle": "space", "indentWidth": 2 } }, { "includes": " / .d.ts" , "linter": { "rules": { "complexity": { "noBannedTypes": "off" }… 证据：`biome.json`
- **Tsconfig**（structured_config）：{ "compilerOptions": { "target": "ES2022", "module": "ESNext", "moduleResolution": "bundler", "strict": true, "esModuleInterop": true, "skipLibCheck": true, "outDir": "dist", "rootDir": "src", "declaration": false }, "include": "src" } 证据：`mcp-bridge/tsconfig.json`
- **Tsconfig.Base**（structured_config）：{ "compilerOptions": { "strict": true, "target": "esnext", "module": "esnext", "moduleResolution": "bundler", "baseUrl": ".", "paths": { "@aipexstudio/aipex-core": "./packages/core/src" , "@aipexstudio/aipex-core/ ": "./packages/core/src/ " , "@aipexstudio/aipex-react": "./packages/aipex-react/src" , "@aipexstudio/aipex-react/ ": "./packages/aipex-react/src/ " , "@aipexstudio/browser-runtime": "./packages/browser-runtime/src" , "@aipexstudio/browser-runtime/ ": "./packages/browser-runtime/src/ " , "@aipexstudio/dom-snapshot": "./packages/dom-snapshot/src" , "@aipexstudio/dom-snapshot/ ": "./packages/dom-snapshot/src/ " }, "downlevelIteration": true, "esModuleInterop": true, "skipLibCheck":… 证据：`tsconfig.base.json`
- **Tsconfig**（structured_config）：{ "extends": "./tsconfig.base.json", "include": "./ .ts", "./ .tsx", "./ .js", "./ .cjs", "./ .mjs", "./types/ / .d.ts" , "references": { "path": "./packages/core" }, { "path": "./packages/browser-runtime" }, { "path": "./packages/aipex-react" }, { "path": "./packages/browser-ext" } } 证据：`tsconfig.json`
- **En**（structured_config）：{ "common": { "title": "AI Chat", "settings": "Settings", "newChat": "New Chat", "close": "Close", "save": "Save Settings", "saving": "Saving...", "cancel": "Cancel", "send": "Send", "stop": "Stop", "processing": "Processing...", "noActions": "No actions", "showThinkingDetails": "Show thinking details", "clickToExpand": "Click to expand" }, "settings": { "title": "Settings", "subtitle": "AI configuration", "language": "Language", "theme": "Theme", "byok": "BYOK Bring Your Own Key ", "byokDescription": "Enable to use your own AI API key and configuration", "aiHost": "AI Host", "aiToken": "API Key", "aiModel": "AI Model", "hostPlaceholder": "https://api.deepseek.com/v1", "tokenPlaceholder": "… 证据：`packages/aipex-react/src/i18n/locales/en.json`
- **Zh**（structured_config）：{ "common": { "title": "AI 对话", "settings": "设置", "newChat": "新对话", "close": "关闭", "save": "保存设置", "saving": "保存中...", "cancel": "取消", "send": "发送", "stop": "停止", "processing": "处理中...", "noActions": "无可用操作", "showThinkingDetails": "显示思考过程", "clickToExpand": "点击展开" }, "settings": { "title": "设置", "subtitle": "AI 配置", "language": "语言", "theme": "主题", "byok": "BYOK 使用您自己的密钥 ", "byokDescription": "启用以使用您自己的 AI API 密钥和配置", "aiHost": "AI 服务地址", "aiToken": "API 密钥", "aiModel": "AI 模型", "hostPlaceholder": "https://api.deepseek.com/v1", "tokenPlaceholder": "请输入您的 API 令牌", "modelPlaceholder": "deepseek-chat", "saveSuccess": "设置保存成功！", "saveError": "保存设置时出错", "testConnection": "测试连接", "testing": "测试中… 证据：`packages/aipex-react/src/i18n/locales/zh.json`
- **Tsconfig**（structured_config）：{ "extends": "../../tsconfig.base.json", "compilerOptions": { "jsx": "react-jsx", "lib": "esnext", "dom", "dom.iterable" , "types": "node", "chrome" , "emitDeclarationOnly": false, "outDir": "./dist" }, "include": "src/ / ", "vitest.setup.ts", "src/i18n/locales/ .json" , "references": { "path": "../core/tsconfig.json" } } 证据：`packages/aipex-react/tsconfig.json`
- **Components**（structured_config）：{ "$schema": "https://ui.shadcn.com/schema.json", "style": "new-york", "rsc": false, "tsx": true, "tailwind": { "config": "tailwind.config.js", "css": "src/tailwind.css", "baseColor": "neutral", "cssVariables": true, "prefix": "" }, "iconLibrary": "lucide" } 证据：`packages/browser-ext/components.json`
- 其余 15 条证据见 `AI_CONTEXT_PACK.json` 或 `EVIDENCE_INDEX.json`。

## 宿主 AI 必须遵守的规则

- **把本资产当作开工前上下文，而不是运行环境。**：AI Context Pack 只包含证据化项目理解，不包含目标项目的可执行状态。 证据：`README.md`, `mcp-bridge/README.md`, `packages/aipex-react/README.md`
- **回答用户时区分可预览内容与必须安装后才能验证的内容。**：安装前体验的消费者价值来自降低误装和误判，而不是伪装成真实运行。 证据：`README.md`, `mcp-bridge/README.md`, `packages/aipex-react/README.md`

## 用户开工前应该回答的问题

- 你准备在哪个宿主 AI 或本地环境中使用它？
- 你只是想先体验工作流，还是准备真实安装？
- 你最在意的是安装成本、输出质量、还是和现有规则的冲突？

## 验收标准

- 所有能力声明都能回指到 evidence_refs 中的文件路径。
- AI_CONTEXT_PACK.md 没有把预览包装成真实运行。
- 用户能在 3 分钟内看懂适合谁、能做什么、如何开始和风险边界。

---

## Doramagic Context Augmentation

下面内容用于强化 Repomix/AI Context Pack 主体。Human Manual 只提供阅读骨架；踩坑日志会被转成宿主 AI 必须遵守的工作约束。

## Human Manual 骨架

使用规则：这里只是项目阅读路线和显著性信号，不是事实权威。具体事实仍必须回到 repo evidence / Claim Graph。

宿主 AI 硬性规则：
- 不得把页标题、章节顺序、摘要或 importance 当作项目事实证据。
- 解释 Human Manual 骨架时，必须明确说它只是阅读路线/显著性信号。
- 能力、安装、兼容性、运行状态和风险判断必须引用 repo evidence、source path 或 Claim Graph。

- **AIPex 系统架构总览**：importance `high`
  - source_paths: packages/core/src/agent/aipex.ts, packages/core/src/index.ts, packages/browser-runtime/src/index.ts, packages/browser-ext/manifest.json, packages/browser-runtime/src/runtime/browser-automation-host.ts
- **浏览器自动化引擎：DOM 快照、定位器与工具集**：importance `high`
  - source_paths: packages/dom-snapshot/src/collector.ts, packages/dom-snapshot/src/manager.ts, packages/dom-snapshot/src/query.ts, packages/dom-snapshot/src/snapshot-formatter.ts, packages/browser-runtime/src/automation/smart-locator.ts
- **MCP 桥接与 AI 代理集成**：importance `high`
  - source_paths: mcp-bridge/src/bridge.ts, mcp-bridge/src/cli.ts, mcp-bridge/src/daemon.ts, mcp-bridge/src/tool-schemas.ts, packages/browser-runtime/src/ws-bridge/ws-mcp-server.ts
- **技能系统、对话管理与用户界面**：importance `medium`
  - source_paths: packages/browser-runtime/src/skill/lib/services/skill-manager.ts, packages/browser-runtime/src/skill/lib/services/skill-registry.ts, packages/browser-runtime/src/skill/lib/services/skill-executor.ts, packages/browser-runtime/src/skill/lib/storage/skill-storage.ts, packages/browser-runtime/src/lib/vm/quickjs-manager.ts

## Repo Inspection Evidence / 源码检查证据

- repo_clone_verified: true
- repo_inspection_verified: true
- repo_commit: `0ad38bf820af48ec06eba06cfc97eab347880bfa`
- inspected_files: `README.md`, `package.json`, `pnpm-lock.yaml`, `packages/aipex-react/README.md`, `packages/aipex-react/package.json`, `packages/aipex-react/src/adapters/chat-adapter.test.ts`, `packages/aipex-react/src/adapters/chat-adapter.ts`, `packages/aipex-react/src/components/chatbot/components/index.ts`, `packages/aipex-react/src/components/chatbot/components/slots/index.ts`, `packages/aipex-react/src/components/chatbot/constants.ts`, `packages/aipex-react/src/components/chatbot/context.ts`, `packages/aipex-react/src/components/chatbot/index.ts`, `packages/aipex-react/src/components/chatbot/themes.ts`, `packages/aipex-react/src/components/chatbot/tools.ts`, `packages/aipex-react/src/components/error/index.ts`, `packages/aipex-react/src/components/fake-mouse/types.ts`, `packages/aipex-react/src/components/file-manager/index.ts`, `packages/aipex-react/src/components/file-manager/types.ts`, `packages/aipex-react/src/components/file-manager/utils.ts`, `packages/aipex-react/src/components/intervention/index.ts`

宿主 AI 硬性规则：
- 没有 repo_clone_verified=true 时，不得声称已经读过源码。
- 没有 repo_inspection_verified=true 时，不得把 README/docs/package 文件判断写成事实。
- 没有 quick_start_verified=true 时，不得声称 Quick Start 已跑通。

## Doramagic Pitfall Constraints / 踩坑约束

这些规则来自 Doramagic 发现、验证或编译过程中的项目专属坑点。宿主 AI 必须把它们当作工作约束，而不是普通说明文字。

### Constraint 1: 涉及密钥、隐私或敏感领域

- Trigger: 项目文本出现 secret/private key/privacy/trading/finance 等敏感关键词。
- Host AI rule: 补敏感数据流、密钥存储和权限边界审查。
- Why it matters: 金融、交易、隐私和密钥场景必须比普通工具更保守。
- Evidence: packet_text.keyword_scan | https://github.com/AIPexStudio/AIPex | matched secret / private key / privacy / trading / finance keyword
- Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。

### Constraint 2: 可能修改宿主 AI 配置

- Trigger: 项目面向 Claude/Cursor/Codex/Gemini/OpenCode 等宿主，或安装命令涉及用户配置目录。
- Host AI rule: 列出会写入的配置文件、目录和卸载/回滚步骤。
- Why it matters: 安装可能改变本机 AI 工具行为，用户需要知道写入位置和回滚方法。
- Evidence: capability.host_targets | https://github.com/AIPexStudio/AIPex | host_targets=claude, chatgpt, mcp_host, claude_code, cursor
- Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。

### Constraint 3: 能力判断依赖假设

- Trigger: README/documentation is current enough for a first validation pass.
- Host AI rule: 将假设转成下游验证清单。
- Why it matters: 假设不成立时，用户拿不到承诺的能力。
- Evidence: capability.assumptions | https://github.com/AIPexStudio/AIPex | README/documentation is current enough for a first validation pass.
- Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。

### Constraint 4: 维护活跃度未知

- Trigger: 未记录 last_activity_observed。
- Host AI rule: 补 GitHub 最近 commit、release、issue/PR 响应信号。
- Why it matters: 新项目、停更项目和活跃项目会被混在一起，推荐信任度下降。
- Evidence: evidence.maintainer_signals | https://github.com/AIPexStudio/AIPex | last_activity_observed missing
- Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。

- Trigger: no_demo
- Evidence: downstream_validation.risk_items | https://github.com/AIPexStudio/AIPex | no_demo; severity=medium
- Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。

### Constraint 6: 存在评分风险

- Trigger: no_demo
- Why it matters: 风险会影响是否适合普通用户安装。
- Evidence: risks.scoring_risks | https://github.com/AIPexStudio/AIPex | no_demo; severity=medium
- Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。

### Constraint 7: issue/PR 响应质量未知

- Trigger: issue_or_pr_quality=unknown。
- Host AI rule: 抽样最近 issue/PR，判断是否长期无人处理。
- Why it matters: 用户无法判断遇到问题后是否有人维护。
- Evidence: evidence.maintainer_signals | https://github.com/AIPexStudio/AIPex | issue_or_pr_quality=unknown
- Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。

### Constraint 8: 发布节奏不明确

- Trigger: release_recency=unknown。
- Host AI rule: 确认最近 release/tag 和 README 安装命令是否一致。
- Why it matters: 安装命令和文档可能落后于代码，用户踩坑概率升高。
- Evidence: evidence.maintainer_signals | https://github.com/AIPexStudio/AIPex | release_recency=unknown
- Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。