# potpie - 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 potpie 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_0004` 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：`potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-change-timeline/SKILL.md`, `potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-cli/SKILL.md`, `potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-debug-memory/SKILL.md`, `potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-graph/SKILL.md` 等 Claim：`clm_0005` 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：`potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-change-timeline/SKILL.md`, `potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-cli/SKILL.md`, `potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-debug-memory/SKILL.md`, `potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-graph/SKILL.md` 等 Claim：`clm_0001` supported 0.86
- **多宿主安装与分发**（需要安装后验证）：项目包含插件或 marketplace 配置，说明它面向一个或多个 AI 宿主的安装和分发。 证据：`potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/.claude-plugin/marketplace.json`, `potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/.claude-plugin/plugin.json` Claim：`clm_0002` 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` Claim：`clm_0003` supported 0.86

## 怎么开始

- `uv tool install potpie` 证据：`README.md` Claim：`clm_0006` supported 0.86

## 继续前判断卡

- **当前建议**：仅建议沙盒试装
- **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**：仅建议沙盒试装
- **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**：真实输出质量不能在安装前相信。
- **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_0004` 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：`potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-change-timeline/SKILL.md`, `potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-cli/SKILL.md`, `potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-debug-memory/SKILL.md`, `potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-graph/SKILL.md` 等 Claim：`clm_0005` 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：`potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-change-timeline/SKILL.md`, `potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-cli/SKILL.md`, `potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-debug-memory/SKILL.md`, `potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-graph/SKILL.md` 等 Claim：`clm_0001` supported 0.86
- **能力存在：多宿主安装与分发**（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：`potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/.claude-plugin/marketplace.json`, `potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/.claude-plugin/plugin.json` Claim：`clm_0002` 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` Claim：`clm_0003` 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_0006` supported 0.86

### 现在还不能相信

- **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：`potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-change-timeline/SKILL.md`, `potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-cli/SKILL.md`, `potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-debug-memory/SKILL.md`, `potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-graph/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。 证据：`potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/.claude-plugin/marketplace.json`, `potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/.claude-plugin/plugin.json`
- **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`
- **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：`potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-change-timeline/SKILL.md`, `potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-cli/SKILL.md`, `potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-debug-memory/SKILL.md`, `potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-graph/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`, `potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/.claude-plugin/marketplace.json`, `potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/.claude-plugin/plugin.json`
- **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_0007` inferred 0.45
- **宿主 AI 插件或 Skill 规则冲突**：新规则可能改变用户现有宿主 AI 的工作方式。 处理方式：安装前先检查插件 manifest 和 Skill 文件，必要时隔离测试。 证据：`potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/.claude-plugin/marketplace.json`, `potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/.claude-plugin/plugin.json` Claim：`clm_0008` supported 0.86
- **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` Claim：`clm_0009` 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：`potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-change-timeline/SKILL.md`, `potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-cli/SKILL.md`, `potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-debug-memory/SKILL.md`, `potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-graph/SKILL.md` 等 Claim：`clm_0001` supported 0.86
- **多宿主安装与分发**：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：`potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/.claude-plugin/marketplace.json`, `potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/.claude-plugin/plugin.json` Claim：`clm_0002` 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` Claim：`clm_0003` supported 0.86

### 上下文规模

- 文件总数：1495
- 重要文件覆盖：40/1495
- 证据索引条目：80
- 角色 / Skill 条目：17

### 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 potpie 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 potpie 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 potpie 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 17 个角色 / Skill / 项目文档条目。

- **potpie-change-timeline**（skill）：Use when an agent needs recent or historical change context: what changed recently, regressions, merged PRs, tickets, docs, incidents, deployments, releases, and source-history ingestion. 激活提示：当用户任务与“potpie-change-timeline”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：`potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-change-timeline/SKILL.md`
- **potpie-cli**（skill）：Use when the task is centered on running, explaining, configuring, or troubleshooting the potpie command: doctor, login, pot management, source registration, search, graph workbench reads/writes, and pot scope behavior. 激活提示：当用户任务与“potpie-cli”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：`potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-cli/SKILL.md`
- **potpie-debug-memory**（skill）：Use while debugging or troubleshooting failures, flaky tests, incidents, production alerts, CI failures, local dev setup issues, repeated bugs, prior fixes, failed attempts, and verification history. 激活提示：当用户任务与“potpie-debug-memory”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：`potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-debug-memory/SKILL.md`
- **potpie-graph**（skill）：Use when the task can read or write the project-memory graph through the potpie CLI: discover the contract with graph catalog , read named views with graph read , resolve entity identity with graph search-entities , create validated plans with graph propose , commit plans with graph commit --verify , inspect quality with graph quality , or capture uncertain work with graph inbox . Also covers writing retrieval-grade… 激活提示：当用户任务与“potpie-graph”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：`potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-graph/SKILL.md`
- **potpie-infra-architecture**（skill）：Use for project infra and architecture context: environments, adapters, runtime configuration, deployments, service dependencies, datastores, API contracts, ownership, incidents, and dependency blast radius. 激活提示：当用户任务与“potpie-infra-architecture”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：`potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-infra-architecture/SKILL.md`
- **potpie-project-preferences**（skill）：Use before writing, modifying, reviewing, refactoring, or testing code so repo/project preferences surface: error handling, file structure, frameworks, logging, dependency choices, testing, security, API style, and naming. Also use after code work when a reusable project preference should be recorded. 激活提示：当用户任务与“potpie-project-preferences”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：`potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-project-preferences/SKILL.md`
- **potpie-repo-baseline**（skill）：Use when establishing, refreshing, or deeply understanding a repository's baseline memory in Potpie: purpose, application type, features, services/modules, environments, deploy shape, dependencies, API contracts, datastores, integrations, ownership, and explicit preferences. The harness reads authored and code-adjacent sources, then writes graph workbench mutations. 激活提示：当用户任务与“potpie-repo-baseline”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：`potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-repo-baseline/SKILL.md`
- **potpie-source-ingestion**（skill）：Use when the user explicitly asks to ingest, refresh, or deeply understand a repository, PR, issue, ticket, runbook, incident report, document, or web link into Potpie. The harness performs todo-driven discovery, uses local/GitHub/integration tools and read-only subagents when available, builds evidence-backed semantic mutations, and writes through graph propose/verified commit. 激活提示：当用户任务与“potpie-source-ingestion”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：`potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-source-ingestion/SKILL.md`
- **potpie-change-timeline**（skill）：Use when an agent needs recent or historical change context: what changed recently, regressions, merged PRs, tickets, docs, incidents, deployments, releases, and source-history ingestion. 激活提示：当用户任务与“potpie-change-timeline”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：`potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/skills/potpie-change-timeline/SKILL.md`
- **potpie-debug-memory**（skill）：Use while debugging or troubleshooting failures, flaky tests, incidents, production alerts, CI failures, local dev setup issues, repeated bugs, prior fixes, failed attempts, and verification history. 激活提示：当用户任务与“potpie-debug-memory”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：`potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/skills/potpie-debug-memory/SKILL.md`
- **potpie-graph**（skill）：Use when the task can read or write the project-memory graph through the potpie CLI: discover the contract with graph catalog , read named views with graph read , resolve entity identity with graph search-entities , create validated plans with graph propose , commit plans with graph commit --verify , inspect quality with graph quality , or capture uncertain work with graph inbox . Also covers writing retrieval-grade… 激活提示：当用户任务与“potpie-graph”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：`potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/skills/potpie-graph/SKILL.md`
- **potpie-infra-architecture**（skill）：Use for project infra and architecture context: environments, adapters, runtime configuration, deployments, service dependencies, datastores, API contracts, ownership, incidents, and dependency blast radius. 激活提示：当用户任务与“potpie-infra-architecture”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：`potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/skills/potpie-infra-architecture/SKILL.md`
- **potpie-project-preferences**（skill）：Use before writing, modifying, reviewing, refactoring, or testing code so repo/project preferences surface: error handling, file structure, frameworks, logging, dependency choices, testing, security, API style, and naming. Also use after code work when a reusable project preference should be recorded. 激活提示：当用户任务与“potpie-project-preferences”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：`potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/skills/potpie-project-preferences/SKILL.md`
- **potpie-repo-baseline**（skill）：Use when establishing, refreshing, or deeply understanding a repository's baseline memory in Potpie: purpose, application type, features, services/modules, environments, deploy shape, dependencies, API contracts, datastores, integrations, ownership, and explicit preferences. The harness reads authored and code-adjacent sources, then writes graph workbench mutations. 激活提示：当用户任务与“potpie-repo-baseline”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：`potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/skills/potpie-repo-baseline/SKILL.md`
- **potpie-source-ingestion**（skill）：Use when the user explicitly asks to ingest, refresh, or deeply understand a repository, PR, issue, ticket, runbook, incident report, document, or web link into Potpie. The harness performs todo-driven discovery, uses local/GitHub/integration tools and read-only subagents when available, builds evidence-backed semantic mutations, and writes through graph propose/verified commit. 激活提示：当用户任务与“potpie-source-ingestion”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：`potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/skills/potpie-source-ingestion/SKILL.md`
- **backfill-enumerate-drain**（skill）：How to backfill a newly-added repo or team — the enumerate-then-drain loop, the single-event contract, and resume-safe todo discipline. Load this when handling a .added seed event with the planner on. 激活提示：当用户任务与“backfill-enumerate-drain”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：`potpie/context-engine/adapters/outbound/reconciliation/skills/backfill-enumerate-drain/SKILL.md`
- **graph-mutation-plan**（skill）：Cookbook for composing an apply graph mutations plan — stable entity key patterns, the canonical label/edge vocabulary, evidence/invalidation/confidence discipline, and a worked example. Load this when building a non-trivial mutation plan. 激活提示：当用户任务与“graph-mutation-plan”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：`potpie/context-engine/adapters/outbound/reconciliation/skills/graph-mutation-plan/SKILL.md`

## 证据索引

- 共索引 80 条证据。

- **Context Graph Docs**（documentation）：The Context Graph is Potpie's project-context layer for agents. Users and agents talk to the potpie CLI. Graph V1 is the current implementation target: keep the existing v1 compatibility surface, move intelligence into the user's harness through skills, and adopt the forward-compatible ontology/truth/evidence model. Graph V2 is the later workbench surface that exposes those internals through potpie graph ... . The same Pot Management, Graph Service, and Skill Manager modules run inside either a local daemon or a managed backend API. Graph state stays local by default unless the user logs in to a managed backend and selects a managed pot, or explicitly runs cloud graph sync. 证据：`docs/context-graph/README.md`
- **Install and setup Potpie**（documentation）：Potpie https://potpie.ai turns your codebase and software development lifecycle into a living context graph for AI agents. It indexes code, structure, decisions, source history, team knowledge and engineering workflows, so agents can answer questions, plan changes, debug failures, and write code with project-specific context. 证据：`README.md`
- **Potpie context-graph host app/modules/context graph**（documentation）：Potpie context-graph host app/modules/context graph 证据：`legacy/app/modules/context_graph/README.md`
- **Confluence Tools**（documentation）：A collection of LangChain tools for interacting with Confluence Cloud via the REST API v2. 证据：`legacy/app/modules/intelligence/tools/confluence_tools/README.md`
- **Jira Integration Tools & Agent**（documentation）：This module provides AI agent tools for interacting with Jira using OAuth2 authentication, plus a specialized Jira Integration Agent. 证据：`legacy/app/modules/intelligence/tools/jira_tools/README.md`
- **Linear Integration Tools**（documentation）：This directory contains tools for integrating with the Linear project management system. 证据：`legacy/app/modules/intelligence/tools/linear_tools/README.md`
- **Tool Registry Phase 1 + Phase 2**（documentation）：Single source of truth for tool metadata and agent–tool binding. Agents resolve tool names via allow-lists or categories instead of hardcoded lists. 证据：`legacy/app/modules/intelligence/tools/registry/README.md`
- **Observability deployment — Loki + Promtail + Grafana**（documentation）：Observability deployment — Loki + Promtail + Grafana 证据：`legacy/deploy/observability/README.md`
- **Database seed JSON profiles**（documentation）：Load deterministic demo data into Postgres using the same .env as the app POSTGRES SERVER . 证据：`legacy/seed/README.md`
- **Potpie CLI potpie**（documentation）：The host-routed command-line entrypoint for the context graph. Every command routes CLI → HostShell → service s → ports ; the in-process HostShell is the single composition root for the agent surface shared with the MCP server . 证据：`potpie/context-engine/adapters/inbound/cli/README.md`
- **Potpie plugin for Claude Code**（documentation）：Wires Claude Code lifecycle events to Potpie's project-memory graph so relevant context surfaces without the agent being asked , and durable learnings can be captured at the right moments. The hook path is model-free : it forwards event shape to potpie graph nudge and injects the returned context/instruction. The reasoning - deciding what's true, which entity, and how to phrase it - happens in your session on your subscription. 证据：`potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/README.md`
- **Context Engine Benchmarks**（documentation）：A live, no-mock benchmark harness for the context engine. Every scenario: 证据：`potpie/context-engine/benchmarks/README.md`
- **Fixtures: recorded webhook payloads**（documentation）：Fixtures: recorded webhook payloads 证据：`potpie/context-engine/benchmarks/fixtures/README.md`
- **potpie-integrations**（documentation）：External integrations for Potpie: OAuth providers, project sources , Linear sync, and HTTP routers. 证据：`potpie/integrations/README.md`
- **Parsing**（documentation）：Extract code graphs from repositories. 证据：`potpie/parsing/README.md`
- **Test Suite**（documentation）：All tests live under this directory. See docs/test-suite-and-parsing-plan.md for the full plan. 证据：`legacy/tests/README.md`
- **Context Engine**（documentation）：This project uses Potpie for project memory. Before non-trivial work, read the graph to orient yourself. After work, record durable learnings that should help the next agent. 证据：`potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/AGENTS.md`
- **Context Engine**（documentation）：This project uses Potpie for project memory. Read the graph before non-trivial work; record durable learnings after. The harness is the intelligence. Potpie validates, lowers, commits, audits, and ranks graph memory. It does not scan the repository or decide what prose means for you. 证据：`potpie/context-engine/adapters/inbound/cli/templates/claude_bundle/CLAUDE.md`
- **Agents**（documentation）：Potpie is durable project memory: repo/source mappings, decisions, infra, changes, bugs, docs, and preferences for agents. Use it when it can materially help with repo context, prior decisions, architecture, bugs, or durable history. Do not run Potpie checks for simple Q&A or trivial edits. When useful, check mapping/graph health once per session potpie --json source list , potpie --json graph status ; skip if unavailable . Record only durable learnings. 证据：`potpie/context-engine/adapters/inbound/cli/templates/global_agent_bundle/AGENTS.md`
- **Claude**（documentation）：Potpie is durable project memory: repo/source mappings, decisions, infra, changes, bugs, docs, and preferences for agents. Use it when it can materially help with repo context, prior decisions, architecture, bugs, or durable history. Do not run Potpie checks for simple Q&A or trivial edits. When useful, check mapping/graph health once per session potpie --json source list , potpie --json graph status ; skip if unavailable . Record only durable learnings. 证据：`potpie/context-engine/adapters/inbound/cli/templates/global_agent_bundle/CLAUDE.md`
- **Package**（package_manifest）：{ "name": "potpie-graph-explorer", "private": true, "version": "0.1.0", "type": "module", "scripts": { "dev": "vite", "build": "tsc -b && vite build", "preview": "vite preview" }, "dependencies": { "react": "^18.3.1", "react-dom": "^18.3.1", "react-force-graph-2d": "^1.27.0" }, "devDependencies": { "@types/react": "^18.3.12", "@types/react-dom": "^18.3.1", "@vitejs/plugin-react": "^4.3.4", "typescript": "^5.6.3", "vite": "^5.4.11" } } 证据：`potpie/context-engine/adapters/inbound/http/ui/frontend/package.json`
- **Potpie Change Timeline**（skill_instruction）：Use this skill when the user asks what changed, when debugging a possible regression, or when ingesting source history from GitHub, Linear, Jira, docs, or deployment records. 证据：`potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-change-timeline/SKILL.md`
- **Potpie CLI**（skill_instruction）：Use this skill when the user is asking about the potpie command itself. For ordinary project-memory context, prefer the relevant use-case skill first. 证据：`potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-cli/SKILL.md`
- **Potpie Debug Memory**（skill_instruction）：Use this skill before digging into a failure so prior symptoms, known fixes, and failed attempts can guide the investigation. 证据：`potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-debug-memory/SKILL.md`
- **Potpie Graph Workbench**（skill_instruction）：The graph is project memory: preferences, prior bugs and their fixes, infra topology, decisions, and a timeline of changes. You are the intelligence that reads it before acting and writes durable learnings after. Potpie validates, lowers, commits, audits, and ranks. It does not scan a repository or infer rich facts from prose for you. 证据：`potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-graph/SKILL.md`
- **Potpie Infra Architecture**（skill_instruction）：Use this skill when the task touches environments, adapters, deployment behavior, runtime config, service dependencies, production incidents, or architecture changes. 证据：`potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-infra-architecture/SKILL.md`
- **Potpie Project Preferences**（skill_instruction）：Use this skill before implementation or review so local conventions shape the work instead of being rediscovered from code. 证据：`potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-project-preferences/SKILL.md`
- **Potpie Repo Baseline**（skill_instruction）：Use this skill when the user asks to ingest, refresh, establish, or deeply understand what a repository is and how it works. 证据：`potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-repo-baseline/SKILL.md`
- **Potpie Source Ingestion**（skill_instruction）：Use this skill for explicit ingestion requests. The harness is the intelligence: it gathers source data, reads it, decides what is durable, resolves identity, and writes semantic graph mutations with evidence. Potpie validates and stores; it does not decide what source material means. 证据：`potpie/context-engine/adapters/inbound/cli/templates/agent_bundle/.agents/skills/potpie-source-ingestion/SKILL.md`
- **Potpie Change Timeline**（skill_instruction）：Use this skill when the user asks what changed, when debugging a possible regression, or when ingesting source history from GitHub, Linear, Jira, docs, or deployment records. 证据：`potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/skills/potpie-change-timeline/SKILL.md`
- **Potpie Debug Memory**（skill_instruction）：Use this skill before digging into a failure so prior symptoms, known fixes, and failed attempts can guide the investigation. 证据：`potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/skills/potpie-debug-memory/SKILL.md`
- **Potpie Graph Workbench**（skill_instruction）：The graph is project memory: preferences, prior bugs and their fixes, infra topology, decisions, and a timeline of changes. You are the intelligence that reads it before acting and writes durable learnings after. Potpie validates, lowers, commits, audits, and ranks. It does not scan a repository or infer rich facts from prose for you. 证据：`potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/skills/potpie-graph/SKILL.md`
- **Potpie Infra Architecture**（skill_instruction）：Use this skill when the task touches environments, adapters, deployment behavior, runtime config, service dependencies, production incidents, or architecture changes. 证据：`potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/skills/potpie-infra-architecture/SKILL.md`
- **Potpie Project Preferences**（skill_instruction）：Use this skill before implementation or review so local conventions shape the work instead of being rediscovered from code. 证据：`potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/skills/potpie-project-preferences/SKILL.md`
- **Potpie Repo Baseline**（skill_instruction）：Use this skill when the user asks to ingest, refresh, establish, or deeply understand what a repository is and how it works. 证据：`potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/skills/potpie-repo-baseline/SKILL.md`
- **Potpie Source Ingestion**（skill_instruction）：Use this skill for explicit ingestion requests. The harness is the intelligence: it gathers source data, reads it, decides what is durable, resolves identity, and writes semantic graph mutations with evidence. Potpie validates and stores; it does not decide what source material means. 证据：`potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/skills/potpie-source-ingestion/SKILL.md`
- **Backfill: enumerate, then drain**（skill_instruction）：You are handling a single .added seed event a repository or a team was just connected . The graph has no history for it yet. One seed fans out into many artifacts PRs, issues, projects, documents . Your todo/plan tools are ON. Use this procedure for any such backfill; the playbook tells you WHICH enumerators and WHAT to extract per artifact. 证据：`potpie/context-engine/adapters/outbound/reconciliation/skills/backfill-enumerate-drain/SKILL.md`
- **Composing an apply graph mutations plan**（skill_instruction）：Composing an apply graph mutations plan 证据：`potpie/context-engine/adapters/outbound/reconciliation/skills/graph-mutation-plan/SKILL.md`
- **Marketplace**（structured_config）：{ "name": "potpie", "owner": { "name": "Potpie" }, "plugins": { "name": "potpie", "source": "./", "description": "Project-memory context engine: model-free nudges surface graph context, while the in-session harness decides and records durable semantic memory." } } 证据：`potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/.claude-plugin/marketplace.json`
- **Plugin**（structured_config）：{ "name": "potpie", "description": "Project-memory context engine for Claude Code: surfaces preferences, prior bugs/fixes, infra topology, timeline context, and durable-capture prompts. Hooks are model-free; the in-session harness decides what is true and writes semantic graph mutations.", "version": "0.1.0", "author": { "name": "Potpie" }, "keywords": "context", "memory", "knowledge-graph", "potpie", "hooks" , "hooks": "./hooks/hooks.json" } 证据：`potpie/context-engine/adapters/inbound/cli/templates/claude_plugin/.claude-plugin/plugin.json`
- **License**（source_file）：Apache License Version 2.0, January 2004 http://www.apache.org/licenses/ 证据：`LICENSE`
- **Async Migration Plan — Native Libraries, Step by Step**（documentation）：Async Migration Plan — Native Libraries, Step by Step 证据：`docs/async-migration-plan.md`
- **Sandbox Core Setup**（documentation）：Potpie agents need an isolated place to inspect and change user repositories: clone once, reuse the repo later, create worktrees for new work, run commands, edit files, run tests, and resume the same state in a later agent session. 证据：`docs/sandbox-core-setup.md`
- **Sandbox Integration Plan**（documentation）：This document plans the next phase of app/src/sandbox : turning it into the single library that everything else in potpie uses to materialize repos and run code, and retiring the app/modules/repo manager + app/modules/intelligence/ tools/code changes manager stack. 证据：`docs/sandbox-integration-plan.md`
- **VS Code Extension Debug/DAP Handoff**（documentation）：VS Code Extension Debug/DAP Handoff 证据：`docs/vscode-extension-debug-handoff.md`
- **Agent Surface Ingestion And Query Review**（documentation）：Agent Surface Ingestion And Query Review 证据：`docs/context-graph/agent-surface-ingestion-query-review.md`
- **Context Graph Architecture**（documentation）：This document is the implementation map for the Context Graph. It follows the same shape as the product: CLI first, the same service modules hosted either by a local daemon or managed backend API, swappable storage backends, explicit managed-backend login, and explicit cloud sync or ledger commands. 证据：`docs/context-graph/architecture.md`
- **Context Graph Benchmarks**（documentation）：The benchmark validates graph quality. It is not the OSS packaging roadmap. The same scenarios should run against in-memory, embedded, Neo4j, and managed backends through the GraphBackend contract. Source events can enter through explicit harness-style mutation fixtures or an Event Ledger replay path. 证据：`docs/context-graph/bench-plan.md`
- **Potpie CLI Flow And Command Contract**（documentation）：Potpie CLI Flow And Command Contract 证据：`docs/context-graph/cli-flow.md`
- **Graph V2 Workbench Architecture**（documentation）：This document scopes a local-first graph surface for agent harnesses such as Codex, Claude Code, OpenCode, and similar tools. The intelligence lives in the user harness. Potpie provides the CLI, skills, graph contracts, validation, and storage so the agent can query and update project memory directly without learning the whole ontology at once. 证据：`docs/context-graph/graphv2.md`
- **Context Engine — trackable issue candidates pre-PR backlog**（documentation）：Context Engine — trackable issue candidates pre-PR backlog 证据：`docs/context-graph/issue-candidates.md`
- **Observability**（documentation）：Observability must work locally without making OSS installs heavy, and it must scale up to hosted tracing, metrics, logs, readiness, and cost telemetry in managed graph and Event Ledger deployments. 证据：`docs/context-graph/observability.md`
- **PR-split plan — feat/graph-updates → main**（documentation）：PR-split plan — feat/graph-updates → main 证据：`docs/context-graph/pr-split-plan.md`
- **Context Graph Vision**（documentation）：The Context Graph is the shared project memory agents use before, during, and after work. It stores compact, sourced facts about a project so agents do not rebuild context from raw code, PRs, tickets, docs, and chat every time. 证据：`docs/context-graph/vision.md`
- **Graph Workbench And Ontology**（documentation）：This document is the detailed Graph V2 contract. graphv2.md explains the architecture and direction; this file defines the workbench surface, seed ontology, read views, semantic mutations, and the rules for evolving ontology over time. 证据：`docs/context-graph/workbench-ontology.md`
- **Context-Engine CLI Sentry Telemetry**（documentation）：Context-Engine CLI Sentry Telemetry 证据：`docs/telemetry/sentry.md`
- **OAuth 2.0 3LO requires using api.atlassian.com with cloud ID**（source_file）：logger = get logger name ⋮---- async def check confluence integration exists user id: str, db: Session - bool ⋮---- integration = ⋮---- class ConfluenceClient ⋮---- """Client for interacting with Confluence Cloud using OAuth 2.0 3LO authentication.""" def init self, server: str, access token: str, cloud id: str ⋮---- """ Initialize the Confluence client. Args: server: The Confluence server URL e.g., https://yoursite.atlassian.net access token: OAuth 2.0 3LO access token cloud id: Atlassian cloud ID site ID - required for OAuth 2.0 API calls """ ⋮---- OAuth 2.0 3LO requires using api.atlassian.com with cloud ID Using Confluence REST API v2 See: https://developer.atlassian.com/cloud/confluenc… 证据：`legacy/app/modules/intelligence/tools/confluence_tools/confluence_client.py`
- **OAuth 2.0 3LO requires using api.atlassian.com with cloud ID**（source_file）：logger = get logger name ⋮---- async def check jira integration exists user id: str, db: Session - bool ⋮---- integration = ⋮---- class JiraClient ⋮---- """Client for interacting with Jira Cloud using OAuth 2.0 3LO authentication.""" def init self, server: str, access token: str, cloud id: str ⋮---- """ Initialize the Jira client. Args: server: The Jira server URL e.g., https://yoursite.atlassian.net - used for constructing issue URLs access token: OAuth 2.0 3LO access token cloud id: Atlassian cloud ID site ID - required for OAuth 2.0 API calls """ ⋮---- OAuth 2.0 3LO requires using api.atlassian.com with cloud ID See: https://developer.atlassian.com/cloud/jira/platform/oauth-2-3lo-apps/ ⋮… 证据：`legacy/app/modules/intelligence/tools/jira_tools/jira_client.py`
- **Linear Client**（source_file）：LINEAR REQUEST TIMEOUT = httpx.Timeout 30.0, connect=10.0, read=30.0 class LinearClient ⋮---- API URL = "https://api.linear.app/graphql" def init self, api key: str ⋮---- payload = {"query": query} ⋮---- response = client.post self.API URL, headers=self.headers, json=payload ⋮---- result = response.json ⋮---- response = await client.post ⋮---- def get issue self, issue id: str - Dict str, Any ⋮---- variables = {"id": issue id} result = self.execute query self. get issue query , variables ⋮---- async def get issue async self, issue id: str - Dict str, Any ⋮---- result = await self.execute query async self. get issue query , variables ⋮---- @staticmethod def get issue query - str def update i… 证据：`legacy/app/modules/intelligence/tools/linear_tools/linear_client.py`
- **No conversation context — fall back to the user-provided branch.**（source_file）：client: SandboxClient None = None def get sandbox client - SandboxClient ⋮---- settings = settings from env platform provider = GitHubGitPlatformProvider container = build sandbox container client = SandboxClient.from container container ⋮---- def set sandbox client client: SandboxClient None - None ⋮---- client = client ⋮---- AuthKind = Literal "context", "app", "user oauth", "env", "none" ⋮---- @dataclass frozen=True class ResolvedAuth ⋮---- token: str None kind: AuthKind def mint scoped installation token repo name: str - str None ⋮---- app id = os.getenv "GITHUB APP ID" private key = config provider.get github key ⋮---- private key = jwt token = AppAuth app id=app id, private key=privat… 证据：`legacy/app/modules/intelligence/tools/sandbox/client.py`
- 其余 20 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：`docs/context-graph/README.md`, `README.md`, `legacy/app/modules/context_graph/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：`docs/context-graph/README.md`, `README.md`, `legacy/app/modules/context_graph/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。

- **Potpie 概述与快速开始**：importance `high`
  - source_paths: README.md, pyproject.toml, potpie/context-engine/adapters/inbound/cli/host_cli.py, potpie/context-engine/adapters/inbound/cli/commands/bootstrap.py, potpie/context-engine/adapters/inbound/cli/commands/auth.py
- **上下文引擎与 CLI 架构**：importance `high`
  - source_paths: docs/context-graph/architecture.md, potpie/context-engine/bootstrap/container.py, potpie/context-engine/host/daemon.py, potpie/context-engine/adapters/inbound/cli/host_cli.py, potpie/context-engine/application/services/setup_orchestrator.py
- **集成、认证与编码工具适配**：importance `high`
  - source_paths: potpie/context-engine/adapters/outbound/cli_auth/atlassian_client.py, potpie/context-engine/adapters/inbound/cli/auth/atlassian_auth.py, potpie/context-engine/adapters/inbound/cli/auth/github_commands.py, potpie/context-engine/adapters/outbound/connectors/github/connector.py, potpie/integrations/integrations/adapters/outbound/linear/adapter.py
- **沙箱、解析引擎与常见故障**：importance `high`
  - source_paths: potpie/sandbox/sandbox/adapters/outbound/daytona/provider.py, potpie/sandbox/sandbox/adapters/outbound/local/subprocess_runtime.py, potpie/parsing/src/lib.rs, potpie/parsing/src/fff_search/content_index.rs, potpie/context-engine/adapters/outbound/intelligence/local_embedder.py

## Repo Inspection Evidence / 源码检查证据

- repo_clone_verified: true
- repo_inspection_verified: true
- repo_commit: `8dd175bcabd3f7a1a8eef386fef734d6f2cefc31`
- inspected_files: `README.md`, `pyproject.toml`, `uv.lock`, `docs/async-migration-plan.md`, `docs/context-graph/README.md`, `docs/context-graph/agent-surface-ingestion-query-review.md`, `docs/context-graph/architecture.md`, `docs/context-graph/bench-plan.md`, `docs/context-graph/cli-flow.md`, `docs/context-graph/graphv2.md`, `docs/context-graph/issue-candidates.md`, `docs/context-graph/observability.md`, `docs/context-graph/pr-split-plan.md`, `docs/context-graph/vision.md`, `docs/context-graph/workbench-ontology.md`, `docs/sandbox-core-setup.md`, `docs/sandbox-integration-plan.md`, `docs/telemetry/sentry.md`, `docs/vscode-extension-debug-handoff.md`

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: 来源证据：Create shared environment resolution for CLI, daemon, and child processes

- Trigger: GitHub 社区证据显示该项目存在一个安装相关的待验证问题：Create shared environment resolution for CLI, daemon, and child processes
- Why it matters: 可能增加新用户试用和生产接入成本。
- Evidence: community_evidence:github | https://github.com/potpie-ai/potpie/issues/946 | 来源类型 github_issue 暴露的待验证使用条件。
- 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: 来源证据：Add root CLI version command

- Trigger: GitHub 社区证据显示该项目存在一个安装相关的待验证问题：Add root CLI version command
- Host AI rule: 来源显示可能已有修复、规避或版本变化，说明书中必须标注适用版本。
- Why it matters: 可能增加新用户试用和生产接入成本。
- Evidence: community_evidence:github | https://github.com/potpie-ai/potpie/issues/924 | 来源讨论提到 python 相关条件，需在安装/试用前复核。
- 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: 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/potpie-ai/potpie | host_targets=codex, chatgpt
- 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: 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/potpie-ai/potpie | 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 5: 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/potpie-ai/potpie | 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/potpie-ai/potpie | 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: 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/potpie-ai/potpie | 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 8: 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/potpie-ai/potpie | 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 9: 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/potpie-ai/potpie | 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。