# eddi - 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 eddi 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_0002` supported 0.86

## 它能做什么

- **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`, `docs/developer-quickstart.md` Claim：`clm_0001` supported 0.86

## 怎么开始

- `curl -fsSL https://raw.githubusercontent.com/labsai/EDDI/main/install.sh | bash` 证据：`README.md` Claim：`clm_0003` supported 0.86
- `git clone https://github.com/labsai/EDDI.git` 证据：`docs/developer-quickstart.md` Claim：`clm_0004` supported 0.86
- `curl -X POST http://localhost:7070/regulardictionarystore/regulardictionaries \` 证据：`docs/developer-quickstart.md` Claim：`clm_0005` supported 0.86
- `curl -X POST http://localhost:7070/behaviorstore/behaviorsets \` 证据：`docs/developer-quickstart.md` Claim：`clm_0006` supported 0.86
- `curl -X POST http://localhost:7070/outputstore/outputsets \` 证据：`docs/developer-quickstart.md` Claim：`clm_0007` supported 0.86
- `curl -X POST http://localhost:7070/packagestore/packages \` 证据：`docs/developer-quickstart.md` Claim：`clm_0008` supported 0.86
- `curl -X POST http://localhost:7070/agentstore/agents \` 证据：`docs/developer-quickstart.md` Claim：`clm_0009` supported 0.86
- `curl -X POST "http://localhost:7070/administration/production/deploy/agent-abc-123?version=1"` 证据：`docs/developer-quickstart.md` Claim：`clm_0010` supported 0.86
- `curl -X POST http://localhost:7070/agents/agent-abc-123/start \` 证据：`docs/developer-quickstart.md` Claim：`clm_0011` supported 0.86
- `curl -X POST http://localhost:7070/agents/conv-123 \` 证据：`docs/developer-quickstart.md` Claim：`clm_0012` supported 0.86

## 继续前判断卡

- **当前建议**：先做角色匹配试用
- **为什么**：这个项目更像角色库，核心风险是选错角色或把角色文案当执行能力；先用 Prompt Preview 试角色匹配，再决定是否沙盒导入。

### 30 秒判断

- **现在怎么做**：先做角色匹配试用
- **最小安全下一步**：先用 Prompt Preview 试角色匹配；满意后再隔离导入
- **先别相信**：角色质量和任务匹配不能直接相信。
- **继续会触碰**：角色选择偏差、命令执行、宿主 AI 配置

### 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_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`, `docs/developer-quickstart.md` Claim：`clm_0001` 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_0003` supported 0.86

### 现在还不能相信

- **角色质量和任务匹配不能直接相信。**（unverified）：角色库证明有很多角色，不证明每个角色都适合你的具体任务，也不证明角色能产生高质量结果。
- **不能把角色文案当成真实执行能力。**（unverified）：安装前只能判断角色描述和任务画像是否匹配，不能证明它能在宿主 AI 里完成任务。
- **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`, `CLAUDE.md`
- **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。

### 继续会触碰什么

- **角色选择偏差**：用户对任务应该由哪个专家角色处理的判断。 原因：选错角色会让 AI 从错误专业视角回答，浪费时间或误导决策。
- **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`, `docs/developer-quickstart.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：`AGENTS.md`, `CLAUDE.md`
- **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`, `docs/developer-quickstart.md`
- **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**：先用交互式试用验证任务画像和角色匹配，不要先导入整套角色库。（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。
- **保留原始角色选择记录**：如果输出偏题，可以回到任务画像阶段重新选择角色，而不是继续沿着错误角色推进。
- **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`, `docs/developer-quickstart.md` 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

- **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`, `docs/developer-quickstart.md` Claim：`clm_0001` supported 0.86

### 上下文规模

- 文件总数：1849
- 重要文件覆盖：40/1849
- 证据索引条目：80
- 角色 / Skill 条目：74

### 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 eddi 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 eddi 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 eddi 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 74 个角色 / Skill / 项目文档条目。

- **E.D.D.I Documentation**（project_doc）：- 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/README.md`
- **Creating your first Agents**（project_doc）：Prerequisites: Up and Running instance of EDDI see: Getting started ../getting-started.md 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/creating-your-first-agent/README.md`
- **Your first agent**（project_doc）：How to use the godfather agent 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/your-first-agent/README.md`
- **E.D.D.I — Multi-Agent Orchestration Middleware for Conversational AI**（project_doc）：! EDDI Banner Image /screenshots/EDDI-Readme-banner-image.webp 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`README.md`
- **Third-Party Licenses**（project_doc）：This directory contains the license texts for third-party libraries used by EDDI. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`licenses/README.md`
- **EDDI Backend — AI Agent Instructions**（project_doc）：EDDI Backend — AI Agent Instructions 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`AGENTS.md`
- **EDDI — AI Coding Assistant Instructions**（project_doc）：EDDI — AI Coding Assistant Instructions 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`CLAUDE.md`
- **Agent Sync Architecture**（project_doc）：Transport-agnostic pipeline for import, export, upgrade, and live sync of EDDI agent configurations. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/agent-sync-architecture.md`
- **EDDI Architecture**（project_doc）：This document provides a comprehensive overview of EDDI's architecture, design principles, and internal workflow. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/architecture.md`
- **EDDI Ecosystem — Working Changelog**（project_doc）：Purpose: Living document tracking all changes, decisions, and reasoning during implementation. Updated as work progresses for easy reference and review. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/changelog.md`
- **Contributing to EDDI**（project_doc）：Thank you for your interest in contributing to E.D.D.I! This guide will help you get started. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`CONTRIBUTING.md`
- **Table of contents**（project_doc）：- Getting Started getting-started.md - Developer Quickstart Guide developer-quickstart.md - Agent Manager Dashboard agent-manager-gui.md - Creating your first Agent creating-your-first-agent/README.md - Create a "Hello World" agent creating-your-first-agent/creating-your-first-agent.md - Create an agent that reacts to user inputs creating-your-first-agent/creating-your-first-agent-1.md - Your first agent your-first-… 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/SUMMARY.md`
- **Agent-to-Agent A2A Protocol**（project_doc）：Status: Available since EDDI v6.0.0 Spec: Google A2A Protocol https://github.com/google/A2A 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/a2a-protocol.md`
- **Agent Father - New Conversation Flow Diagram**（project_doc）：Agent Father - New Conversation Flow Diagram 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/agent-father-conversation-flow.md`
- **The Agent Father: A Deep Dive**（project_doc）：The Agent Father is EDDI's meta-agent—an agent that creates other agents. It's the perfect example of EDDI's architecture in action, demonstrating how conversation flow, behavior rules, property extraction, and HTTP calls work together to build sophisticated workflows. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/agent-father-deep-dive.md`
- **Agent Father - LangChain Tools Configuration Guide**（project_doc）：Agent Father - LangChain Tools Configuration Guide 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/agent-father-langchain-tools-guide.md`
- **Agent Manager Dashboard**（project_doc）：The EDDI Manager is a modern React 19 single-page application for building, testing, deploying, and monitoring EDDI agents. It is served directly from the EDDI backend — no separate deployment needed. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/agent-manager-gui.md`
- **Agent Sync — Live Instance-to-Instance Synchronization**（project_doc）：Agent Sync — Live Instance-to-Instance Synchronization 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/agent-sync-guide.md`
- **Multimodal Attachments — Usage Guide**（project_doc）：Multimodal Attachments — Usage Guide 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/attachments-guide.md`
- **Audit Ledger**（project_doc）：Status: Available since v6.0.0 EU AI Act: Articles 17/19 — Immutable Decision Traceability 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/audit-ledger.md`
- **Behavior Rules**（project_doc）：Behavior Rules are the decision-making engine in EDDI's Lifecycle Pipeline. They are IF-THEN rules that evaluate conversation state and trigger actions based on conditions. This is where you define when to call an LLM, when to invoke an API, and how your agent responds to user inputs. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/behavior-rules.md`
- **Build Reproducibility**（project_doc）：This document describes what is and is not guaranteed about EDDI builds. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/build-reproducibility.md`
- **CapabilityMatch Condition — Usage Guide**（project_doc）：CapabilityMatch Condition — Usage Guide 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/capability-match-guide.md`
- **Code Review Standards**（project_doc）：This document defines the code review requirements for the EDDI project. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/code-review-standards.md`
- **Compliance Data Flow Diagram**（project_doc）：Audience : Compliance auditors, DPOs, and deployers performing risk assessments. This document provides a single-page overview of how data flows through EDDI, where it's stored, and where encryption is applied. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/compliance-data-flow.md`
- **Conversation Memory and State Management**（project_doc）：Conversation Memory and State Management 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/conversation-memory.md`
- **Conversations**（project_doc）：Conversations are the primary interaction mechanism in EDDI. Each conversation represents a stateful dialog session between a user and an agent, maintaining complete history, context, and state throughout the interaction. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/conversations.md`
- **Deployment Management of Agents**（project_doc）：Deployment Management controls the lifecycle of your agents across different environments. In EDDI, agents go through a create → configure → deploy workflow before they can process conversations. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/deployment-management-of-agents.md`
- **Developer Quickstart Guide**（project_doc）：This guide helps developers quickly understand EDDI's architecture and start building agents. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/developer-quickstart.md`
- **Docker**（project_doc）：This starts EDDI on port 7070 and MongoDB. No login required. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/docker.md`
- **EU AI Act Compliance Guide**（project_doc）：The EU AI Act is the world's first comprehensive AI regulation. It applies to any AI system used within the EU, regardless of where the provider is based. EDDI deployers must classify their agents by risk level and meet corresponding obligations. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/eu-ai-act-compliance.md`
- **Extensions**（project_doc）：Extensions are the building blocks of EDDI agents. In EDDI's composable architecture, agents are not monolithic applications but rather assemblies of extensions , each providing a specific capability. Extensions are referenced by packages, and packages are combined to form complete agents. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/extensions.md`
- **GDPR Compliance Guide**（project_doc）：This guide helps EDDI operators handle data subject requests and meet GDPR/CCPA requirements. For an overview of data processing, see PRIVACY.md ../PRIVACY.md . 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/gdpr-compliance.md`
- **Getting started**（project_doc）：This article will help you to get started with EDDI . 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/getting-started.md`
- **Global Variables**（project_doc）：EDDI includes a built-in global variable store for managing configuration values like default LLM model names, API base URLs, temperature settings, and feature flags. Variables are scoped per tenant — single-tenant deployments use "default" implicitly, while multi-tenant deployments can maintain separate variable sets per tenant. Unlike secrets, global variables are not encrypted and are fully visible in the UI and… 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/global-variables.md`
- **Group Conversations**（project_doc）：Multi-agent structured discussions with moderator synthesis. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/group-conversations.md`
- **HIPAA Compliance Guide**（project_doc）：HIPAA applies when EDDI processes Protected Health Information PHI — for example, when used by healthcare organizations, telehealth platforms, or health-related chatbots. Under HIPAA, EDDI acts as infrastructure used by a Business Associate . 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/hipaa-compliance.md`
- **Human-in-the-Loop HITL Framework**（project_doc）：The HITL framework lets you insert human approval gates into AI agent conversations. It works on both surfaces : 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/hitl.md`
- **FAQs**（project_doc）：1. Create a LangChain configuration with your provider settings 2. Add a behavior rule that triggers the LLM action e.g., send to ai 3. Add the LangChain extension to your package/workflow 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/how-to....md`
- **HttpCalls**（project_doc）：HttpCalls enable EDDI agents to integrate with external REST APIs, making EDDI a powerful orchestration layer that can combine conversational AI with traditional backend services. This is how agents can fetch real-time data, authenticate users, store information in external systems, or trigger business workflows. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/httpcalls.md`
- **Import/Export an Agent**（project_doc）：Import/Export functionality allows you to package entire agents including all their dependencies into portable ZIP files. This is essential for agent lifecycle management, collaboration, and deployment automation. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/import-export-an-agent.md`
- **Data Breach Incident Response Plan**（project_doc）：This runbook outlines the steps for detecting, assessing, and responding to a data breach involving the EDDI platform. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/incident-response.md`
- **Kubernetes Deployment**（project_doc）：EDDI runs natively on Kubernetes. This guide covers deployment options from a simple quickstart to production-grade configurations. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/kubernetes.md`
- **LLM Integration**（project_doc）：The LLM Lifecycle Task formerly "Langchain" is EDDI's unified integration point for Large Language Models LLMs . 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/langchain.md`
- **Log Administration API**（project_doc）：Base path: /administration/logs Security: @RolesAllowed "eddi-admin" — requires eddi-admin role when OIDC is enabled QUARKUS OIDC TENANT ENABLED=true . Bypassed in dev mode OIDC disabled by default . 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/log-administration.md`
- **Managed Agents**（project_doc）：Managed Agents is an EDDI feature that provides automatic conversation management, allowing you to trigger agents based on intents without manually creating and managing conversation IDs. EDDI handles the conversation lifecycle for you. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/managed-agents.md`
- **MCP Server — EDDI as MCP Tool Provider**（project_doc）：MCP Server — EDDI as MCP Tool Provider 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/mcp-server.md`
- **Memory Policy — Commit Flags & Strict Write Discipline**（project_doc）：Memory Policy — Commit Flags & Strict Write Discipline 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/memory-policy.md`
- **Metrics & Monitoring**（project_doc）：E.D.D.I exposes comprehensive metrics via Micrometer https://micrometer.io/ in Prometheus format, covering conversations, tool execution, caching, rate limiting, cost tracking, multi-agent group discussions, scheduled triggers, tenant quotas, audit integrity, and JVM internals. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/metrics.md`
- **Multi-Model Cascading Routing**（project_doc）：Cost-optimized LLM execution via sequential model escalation with confidence-based routing. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/model-cascade.md`
- **Output Configuration**（project_doc）：Output Configurations define what your agent says to users. They are templates that are triggered by actions from Behavior Rules, making them the final step in EDDI's Lifecycle Pipeline. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/output-configuration.md`
- **Output Templating**（project_doc）：Output Templating is one of EDDI's most powerful features—it allows you to create dynamic, data-driven responses that pull information from conversation memory, API responses, or context data. Instead of static text, your agent can generate personalized, contextual replies. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/output-templating.md`
- **Passing Context Information**（project_doc）：Context is external data that you pass from your application into EDDI conversations. It's how you inject real-world information—like user profiles, session data, or business state—into your agent's logic without hard-coding it. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/passing-context-information.md`
- **EDDI Project Philosophy**（project_doc）：The Overarching Directive for All Development in EDDI This document defines the foundational principles that govern every architectural decision, feature implementation, and design trade-off across the EDDI ecosystem. Every contributor — human or AI — must internalize these principles before writing code. This is not a technical specification. Implementation details belong in architecture.md architecture.md and AGEN… 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/project-philosophy.md`
- **Prompt Snippets — Usage Guide**（project_doc）：Prompt Snippets are reusable system prompt building blocks stored as versioned configuration documents. They replace the deleted CounterweightService , IdentityMaskingService , and DeploymentContextService with a flexible, user-extensible, config-driven approach. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/prompt-snippets-guide.md`
- **Properties**（project_doc）：Properties are EDDI's primary mechanism for storing and retrieving state within and across conversations. They are key-value pairs that can be set by agent configuration, extracted from API responses, or written by LLM tools — and they're accessible in every template system prompts, HTTP call bodies, output templates, property instructions . 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/properties.md`
- **Putting It All Together**（project_doc）：This guide shows how all of EDDI's components work together to create a complete, functional agent. We'll build a real-world example step-by-step, explaining how each piece connects. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/putting-it-all-together.md`
- **RAG Retrieval-Augmented Generation**（project_doc）：Phase 8c — Config-driven knowledge base retrieval integrated into the LLM pipeline. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/rag.md`
- **Red Hat Enterprise Linux & OpenShift Support**（project_doc）：Red Hat Enterprise Linux & OpenShift Support 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/redhat-openshift.md`
- **EDDI 6.0.2 Release Notes**（project_doc）：Release Date: April 2026 Platform: Java 25 · Quarkus 3.34.5 · langchain4j 1.13.0 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/release-notes-6.0.2.md`
- **Release Signing & Verification**（project_doc）：Audience: Users, operators, and security auditors who need to verify the integrity of EDDI releases. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/release-signing.md`
- **Release & Versioning Strategy**（project_doc）：Audience: Maintainers, contributors, and CI/CD operators. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/release-versioning.md`
- **Scheduled Execution & Heartbeat Triggers**（project_doc）：Scheduled Execution & Heartbeat Triggers 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/scheduling.md`
- **Secrets Vault**（project_doc）：EDDI includes a built-in secrets vault for managing sensitive values like API keys, tokens, and passwords. Secrets are encrypted at rest, referenced via URI syntax, and automatically scrubbed from logs and API exports. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/secrets-vault.md`
- **Security Review Summary**（project_doc）：This document provides a formal summary of security reviews performed on the EDDI project, as required by the OpenSSF Best Practices https://www.bestpractices.dev/ Gold criteria. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/security-review.md`
- **Security**（project_doc）：This document describes the security measures applied to EDDI's AI Agent Tooling system, particularly for tools that execute in response to LLM-generated arguments, as well as the Keycloak-based authentication layer. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/security.md`
- **Input Pattern Matching for Agent Routing**（project_doc）：Input Pattern Matching for Agent Routing 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/semantic-parser.md`
- **Persistent User Memory**（project_doc）：Persistent User Memory enables EDDI agents to remember facts, preferences, and context about individual users across conversations . Unlike conversation-scoped properties that are lost when a conversation ends, persistent memories survive indefinitely and are automatically loaded into every new conversation with the same user. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/user-memory.md`
- **EDDI Guardrails Architecture — Enterprise Analysis**（project_doc）：EDDI Guardrails Architecture — Enterprise Analysis 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`planning/guardrails-architecture.md`
- **Create an agent that reacts to user inputs**（project_doc）：Create an agent that reacts to user inputs 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/creating-your-first-agent/creating-your-first-agent-1.md`
- **Create a "Hello World" agent**（project_doc）：Prerequisites: Up and Running instance of EDDI see: Getting started ../getting-started.md 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/creating-your-first-agent/creating-your-first-agent.md`
- **EDDI Monitoring Guide**（project_doc）：Enterprise observability for EDDI — metrics, distributed tracing, alerting, and dashboards. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/monitoring/monitoring-guide.md`
- **Business Associate Agreement BAA Template**（project_doc）：Business Associate Agreement BAA Template 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/templates/baa-template.md`
- **Understanding your first agent**（project_doc）：The components of the generated agents 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`docs/your-first-agent/understanding-your-first-agent.md`

## 证据索引

- 共索引 80 条证据。

- **E.D.D.I Documentation**（documentation）：Welcome to the official documentation for E.D.D.I Enhanced Dialog Driven Interface — a production-grade multi-agent orchestration middleware for conversational AI. 证据：`docs/README.md`
- **Creating your first Agents**（documentation）：Prerequisites: Up and Running instance of EDDI see: Getting started ../getting-started.md 证据：`docs/creating-your-first-agent/README.md`
- **Your first agent**（documentation）：After EDDI has started point your browser to this URL: 证据：`docs/your-first-agent/README.md`
- **E.D.D.I — Multi-Agent Orchestration Middleware for Conversational AI**（documentation）：! EDDI Banner Image /screenshots/EDDI-Readme-banner-image.webp 证据：`README.md`
- **Third-Party Licenses**（documentation）：This directory contains the license texts for third-party libraries used by EDDI. 证据：`licenses/README.md`
- **EDDI Backend — AI Agent Instructions**（documentation）：EDDI Backend — AI Agent Instructions 证据：`AGENTS.md`
- **EDDI — AI Coding Assistant Instructions**（documentation）：EDDI — AI Coding Assistant Instructions 证据：`CLAUDE.md`
- **Agent Sync Architecture**（documentation）：Transport-agnostic pipeline for import, export, upgrade, and live sync of EDDI agent configurations. 证据：`docs/agent-sync-architecture.md`
- **EDDI Architecture**（documentation）：This document provides a comprehensive overview of EDDI's architecture, design principles, and internal workflow. 证据：`docs/architecture.md`
- **EDDI Ecosystem — Working Changelog**（documentation）：Purpose: Living document tracking all changes, decisions, and reasoning during implementation. Updated as work progresses for easy reference and review. 证据：`docs/changelog.md`
- **Contributing to EDDI**（documentation）：Thank you for your interest in contributing to E.D.D.I! This guide will help you get started. 证据：`CONTRIBUTING.md`
- **License**（source_file）：Apache License Version 2.0, January 2004 http://www.apache.org/licenses/ 证据：`LICENSE`
- **Table of contents**（documentation）：- Getting Started getting-started.md - Developer Quickstart Guide developer-quickstart.md - Agent Manager Dashboard agent-manager-gui.md - Creating your first Agent creating-your-first-agent/README.md - Create a "Hello World" agent creating-your-first-agent/creating-your-first-agent.md - Create an agent that reacts to user inputs creating-your-first-agent/creating-your-first-agent-1.md - Your first agent your-first-agent/README.md - Understanding your first agent your-first-agent/understanding-your-first-agent.md - Putting It All Together putting-it-all-together.md - Import/Export an Agent import-export-an-agent.md 证据：`docs/SUMMARY.md`
- **Agent-to-Agent A2A Protocol**（documentation）：Status: Available since EDDI v6.0.0 Spec: Google A2A Protocol https://github.com/google/A2A 证据：`docs/a2a-protocol.md`
- **Agent Father - New Conversation Flow Diagram**（documentation）：Agent Father - New Conversation Flow Diagram 证据：`docs/agent-father-conversation-flow.md`
- **The Agent Father: A Deep Dive**（documentation）：The Agent Father is EDDI's meta-agent—an agent that creates other agents. It's the perfect example of EDDI's architecture in action, demonstrating how conversation flow, behavior rules, property extraction, and HTTP calls work together to build sophisticated workflows. 证据：`docs/agent-father-deep-dive.md`
- **Agent Father - LangChain Tools Configuration Guide**（documentation）：Agent Father - LangChain Tools Configuration Guide 证据：`docs/agent-father-langchain-tools-guide.md`
- **Agent Manager Dashboard**（documentation）：The EDDI Manager is a modern React 19 single-page application for building, testing, deploying, and monitoring EDDI agents. It is served directly from the EDDI backend — no separate deployment needed. 证据：`docs/agent-manager-gui.md`
- **Agent Sync — Live Instance-to-Instance Synchronization**（documentation）：Agent Sync — Live Instance-to-Instance Synchronization 证据：`docs/agent-sync-guide.md`
- **Multimodal Attachments — Usage Guide**（documentation）：Multimodal Attachments — Usage Guide 证据：`docs/attachments-guide.md`
- **Audit Ledger**（documentation）：Status: Available since v6.0.0 EU AI Act: Articles 17/19 — Immutable Decision Traceability 证据：`docs/audit-ledger.md`
- **Behavior Rules**（documentation）：Behavior Rules are the decision-making engine in EDDI's Lifecycle Pipeline. They are IF-THEN rules that evaluate conversation state and trigger actions based on conditions. This is where you define when to call an LLM, when to invoke an API, and how your agent responds to user inputs. 证据：`docs/behavior-rules.md`
- **Build Reproducibility**（documentation）：This document describes what is and is not guaranteed about EDDI builds. 证据：`docs/build-reproducibility.md`
- **CapabilityMatch Condition — Usage Guide**（documentation）：CapabilityMatch Condition — Usage Guide 证据：`docs/capability-match-guide.md`
- **Code Review Standards**（documentation）：This document defines the code review requirements for the EDDI project. 证据：`docs/code-review-standards.md`
- **Compliance Data Flow Diagram**（documentation）：Audience : Compliance auditors, DPOs, and deployers performing risk assessments. This document provides a single-page overview of how data flows through EDDI, where it's stored, and where encryption is applied. 证据：`docs/compliance-data-flow.md`
- **Conversation Memory and State Management**（documentation）：Conversation Memory and State Management 证据：`docs/conversation-memory.md`
- **Conversations**（documentation）：Conversations are the primary interaction mechanism in EDDI. Each conversation represents a stateful dialog session between a user and an agent, maintaining complete history, context, and state throughout the interaction. 证据：`docs/conversations.md`
- **Deployment Management of Agents**（documentation）：Deployment Management controls the lifecycle of your agents across different environments. In EDDI, agents go through a create → configure → deploy workflow before they can process conversations. 证据：`docs/deployment-management-of-agents.md`
- **Developer Quickstart Guide**（documentation）：This guide helps developers quickly understand EDDI's architecture and start building agents. 证据：`docs/developer-quickstart.md`
- **Docker**（documentation）：This starts EDDI on port 7070 and MongoDB. No login required. 证据：`docs/docker.md`
- **EU AI Act Compliance Guide**（documentation）：The EU AI Act is the world's first comprehensive AI regulation. It applies to any AI system used within the EU, regardless of where the provider is based. EDDI deployers must classify their agents by risk level and meet corresponding obligations. 证据：`docs/eu-ai-act-compliance.md`
- **Extensions**（documentation）：Extensions are the building blocks of EDDI agents. In EDDI's composable architecture, agents are not monolithic applications but rather assemblies of extensions , each providing a specific capability. Extensions are referenced by packages, and packages are combined to form complete agents. 证据：`docs/extensions.md`
- **GDPR Compliance Guide**（documentation）：This guide helps EDDI operators handle data subject requests and meet GDPR/CCPA requirements. For an overview of data processing, see PRIVACY.md ../PRIVACY.md . 证据：`docs/gdpr-compliance.md`
- **Getting started**（documentation）：This article will help you to get started with EDDI . 证据：`docs/getting-started.md`
- **Global Variables**（documentation）：EDDI includes a built-in global variable store for managing configuration values like default LLM model names, API base URLs, temperature settings, and feature flags. Variables are scoped per tenant — single-tenant deployments use "default" implicitly, while multi-tenant deployments can maintain separate variable sets per tenant. Unlike secrets, global variables are not encrypted and are fully visible in the UI and logs. 证据：`docs/global-variables.md`
- **Group Conversations**（documentation）：Multi-agent structured discussions with moderator synthesis. 证据：`docs/group-conversations.md`
- **HIPAA Compliance Guide**（documentation）：HIPAA applies when EDDI processes Protected Health Information PHI — for example, when used by healthcare organizations, telehealth platforms, or health-related chatbots. Under HIPAA, EDDI acts as infrastructure used by a Business Associate . 证据：`docs/hipaa-compliance.md`
- **Human-in-the-Loop HITL Framework**（documentation）：The HITL framework lets you insert human approval gates into AI agent conversations. It works on both surfaces : 证据：`docs/hitl.md`
- **FAQs**（documentation）：1. Create a LangChain configuration with your provider settings 2. Add a behavior rule that triggers the LLM action e.g., send to ai 3. Add the LangChain extension to your package/workflow 证据：`docs/how-to....md`
- **HttpCalls**（documentation）：HttpCalls enable EDDI agents to integrate with external REST APIs, making EDDI a powerful orchestration layer that can combine conversational AI with traditional backend services. This is how agents can fetch real-time data, authenticate users, store information in external systems, or trigger business workflows. 证据：`docs/httpcalls.md`
- **Import/Export an Agent**（documentation）：Import/Export functionality allows you to package entire agents including all their dependencies into portable ZIP files. This is essential for agent lifecycle management, collaboration, and deployment automation. 证据：`docs/import-export-an-agent.md`
- **Data Breach Incident Response Plan**（documentation）：This runbook outlines the steps for detecting, assessing, and responding to a data breach involving the EDDI platform. 证据：`docs/incident-response.md`
- **Kubernetes Deployment**（documentation）：EDDI runs natively on Kubernetes. This guide covers deployment options from a simple quickstart to production-grade configurations. 证据：`docs/kubernetes.md`
- **LLM Integration**（documentation）：The LLM Lifecycle Task formerly "Langchain" is EDDI's unified integration point for Large Language Models LLMs . 证据：`docs/langchain.md`
- **Log Administration API**（documentation）：Base path: /administration/logs Security: @RolesAllowed "eddi-admin" — requires eddi-admin role when OIDC is enabled QUARKUS OIDC TENANT ENABLED=true . Bypassed in dev mode OIDC disabled by default . 证据：`docs/log-administration.md`
- **Managed Agents**（documentation）：Managed Agents is an EDDI feature that provides automatic conversation management, allowing you to trigger agents based on intents without manually creating and managing conversation IDs. EDDI handles the conversation lifecycle for you. 证据：`docs/managed-agents.md`
- **MCP Server — EDDI as MCP Tool Provider**（documentation）：MCP Server — EDDI as MCP Tool Provider 证据：`docs/mcp-server.md`
- **Memory Policy — Commit Flags & Strict Write Discipline**（documentation）：Memory Policy — Commit Flags & Strict Write Discipline 证据：`docs/memory-policy.md`
- **Metrics & Monitoring**（documentation）：E.D.D.I exposes comprehensive metrics via Micrometer https://micrometer.io/ in Prometheus format, covering conversations, tool execution, caching, rate limiting, cost tracking, multi-agent group discussions, scheduled triggers, tenant quotas, audit integrity, and JVM internals. 证据：`docs/metrics.md`
- **Multi-Model Cascading Routing**（documentation）：Cost-optimized LLM execution via sequential model escalation with confidence-based routing. 证据：`docs/model-cascade.md`
- **Output Configuration**（documentation）：Output Configurations define what your agent says to users. They are templates that are triggered by actions from Behavior Rules, making them the final step in EDDI's Lifecycle Pipeline. 证据：`docs/output-configuration.md`
- **Output Templating**（documentation）：Output Templating is one of EDDI's most powerful features—it allows you to create dynamic, data-driven responses that pull information from conversation memory, API responses, or context data. Instead of static text, your agent can generate personalized, contextual replies. 证据：`docs/output-templating.md`
- **Passing Context Information**（documentation）：Context is external data that you pass from your application into EDDI conversations. It's how you inject real-world information—like user profiles, session data, or business state—into your agent's logic without hard-coding it. 证据：`docs/passing-context-information.md`
- **EDDI Project Philosophy**（documentation）：The Overarching Directive for All Development in EDDI This document defines the foundational principles that govern every architectural decision, feature implementation, and design trade-off across the EDDI ecosystem. Every contributor — human or AI — must internalize these principles before writing code. This is not a technical specification. Implementation details belong in architecture.md architecture.md and AGENTS.md ../AGENTS.md . This document answers why — those documents answer how . 证据：`docs/project-philosophy.md`
- **Prompt Snippets — Usage Guide**（documentation）：Prompt Snippets are reusable system prompt building blocks stored as versioned configuration documents. They replace the deleted CounterweightService , IdentityMaskingService , and DeploymentContextService with a flexible, user-extensible, config-driven approach. 证据：`docs/prompt-snippets-guide.md`
- **Properties**（documentation）：Properties are EDDI's primary mechanism for storing and retrieving state within and across conversations. They are key-value pairs that can be set by agent configuration, extracted from API responses, or written by LLM tools — and they're accessible in every template system prompts, HTTP call bodies, output templates, property instructions . 证据：`docs/properties.md`
- **Putting It All Together**（documentation）：This guide shows how all of EDDI's components work together to create a complete, functional agent. We'll build a real-world example step-by-step, explaining how each piece connects. 证据：`docs/putting-it-all-together.md`
- **RAG Retrieval-Augmented Generation**（documentation）：Phase 8c — Config-driven knowledge base retrieval integrated into the LLM pipeline. 证据：`docs/rag.md`
- **Red Hat Enterprise Linux & OpenShift Support**（documentation）：Red Hat Enterprise Linux & OpenShift Support 证据：`docs/redhat-openshift.md`
- 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/README.md`, `docs/creating-your-first-agent/README.md`, `docs/your-first-agent/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/README.md`, `docs/creating-your-first-agent/README.md`, `docs/your-first-agent/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。

- **项目概览**：importance `high`
  - source_paths: .clusterfuzzlite/Dockerfile, README.md, licenses/README.md, src/main/docker/Dockerfile
- **Api 模块**：importance `high`
  - source_paths: src/main/java/ai/labs/eddi/engine/api/IConversationService.java, src/main/java/ai/labs/eddi/engine/api/IGroupConversationService.java, src/main/java/ai/labs/eddi/engine/api/ILogoutEndpoint.java, src/main/java/ai/labs/eddi/engine/api/IRestAgentAdministration.java, src/main/java/ai/labs/eddi/engine/api/IRestAgentEngine.java
- **Runtime 模块**：importance `high`
  - source_paths: src/main/java/ai/labs/eddi/engine/runtime/BaseRuntime.java, src/main/java/ai/labs/eddi/engine/runtime/BoundedLogStore.java, src/main/java/ai/labs/eddi/engine/runtime/ConversationSetup.java, src/main/java/ai/labs/eddi/engine/runtime/DatabaseLogs.java, src/main/java/ai/labs/eddi/engine/runtime/IAgent.java
- **Client 模块**：importance `high`
  - source_paths: src/main/java/ai/labs/eddi/engine/runtime/client/agents/AgentStoreClientLibrary.java, src/main/java/ai/labs/eddi/engine/runtime/client/agents/IAgentStoreClientLibrary.java, src/main/java/ai/labs/eddi/engine/runtime/client/configuration/IResourceClientLibrary.java, src/main/java/ai/labs/eddi/engine/runtime/client/configuration/ResourceClientLibrary.java, src/main/java/ai/labs/eddi/engine/runtime/client/factory/IRestInterfaceFactory.java
- **Agents 模块**：importance `medium`
  - source_paths: src/main/java/ai/labs/eddi/configs/agents/AgentSigningService.java, src/main/java/ai/labs/eddi/configs/agents/CapabilityRegistryService.java, src/main/java/ai/labs/eddi/configs/agents/IAgentStore.java, src/main/java/ai/labs/eddi/configs/agents/IRestAgentStore.java, src/main/java/ai/labs/eddi/configs/agents/IRestCapabilityRegistry.java
- **Agents 模块**：importance `medium`
  - source_paths: src/main/java/ai/labs/eddi/engine/runtime/client/agents/AgentStoreClientLibrary.java, src/main/java/ai/labs/eddi/engine/runtime/client/agents/IAgentStoreClientLibrary.java
- **Providers 模块**：importance `medium`
  - source_paths: src/main/java/ai/labs/eddi/modules/nlp/extensions/dictionaries/providers/DecimalDictionaryProvider.java, src/main/java/ai/labs/eddi/modules/nlp/extensions/dictionaries/providers/EmailDictionaryProvider.java, src/main/java/ai/labs/eddi/modules/nlp/extensions/dictionaries/providers/IDictionaryProvider.java, src/main/java/ai/labs/eddi/modules/nlp/extensions/dictionaries/providers/IntegerDictionaryProvider.java, src/main/java/ai/labs/eddi/modules/nlp/extensions/dictionaries/providers/OrdinalNumbersDictionaryProvider.java
- **Providers 模块**：importance `medium`
  - source_paths: src/main/java/ai/labs/eddi/modules/nlp/extensions/normalizers/providers/ContractedWordNormalizerProvider.java, src/main/java/ai/labs/eddi/modules/nlp/extensions/normalizers/providers/ConvertSpecialCharacterNormalizerProvider.java, src/main/java/ai/labs/eddi/modules/nlp/extensions/normalizers/providers/INormalizerProvider.java, src/main/java/ai/labs/eddi/modules/nlp/extensions/normalizers/providers/PunctuationNormalizerProvider.java, src/main/java/ai/labs/eddi/modules/nlp/extensions/normalizers/providers/RemoveUndefinedCharacterNormalizerProvider.java

## Repo Inspection Evidence / 源码检查证据

- repo_clone_verified: true
- repo_inspection_verified: true
- repo_commit: `112bb3577232364f686fc1affb6e57422fa9db37`
- inspected_files: `README.md`, `docker-compose.yml`, `docs/README.md`, `docs/SUMMARY.md`, `docs/a2a-protocol.md`, `docs/agent-configs/agent-father/aaa000000000000000000000.agent.json`, `docs/agent-configs/agent-father/aaa000000000000000000000.descriptor.json`, `docs/agent-configs/agent-father/aaa000000000000000000001/1/aaa000000000000000000001.descriptor.json`, `docs/agent-configs/agent-father/aaa000000000000000000001/1/aaa000000000000000000001.workflow.json`, `docs/agent-configs/agent-father/aaa000000000000000000001/1/aaa000000000000000000002.behavior.json`, `docs/agent-configs/agent-father/aaa000000000000000000001/1/aaa000000000000000000002.descriptor.json`, `docs/agent-configs/agent-father/aaa000000000000000000001/1/aaa000000000000000000003.descriptor.json`, `docs/agent-configs/agent-father/aaa000000000000000000001/1/aaa000000000000000000003.property.json`, `docs/agent-configs/agent-father/aaa000000000000000000001/1/aaa000000000000000000004.descriptor.json`, `docs/agent-configs/agent-father/aaa000000000000000000001/1/aaa000000000000000000004.httpcalls.json`, `docs/agent-configs/agent-father/aaa000000000000000000001/1/aaa000000000000000000005.descriptor.json`, `docs/agent-configs/agent-father/aaa000000000000000000001/1/aaa000000000000000000005.output.json`, `docs/agent-father-conversation-flow.md`, `docs/agent-father-deep-dive.md`, `docs/agent-father-langchain-tools-guide.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: 能力证据存在缺口

- Trigger: Sandbox install result is missing.
- Host AI rule: 优先补证或在页面明确标记为未验证。
- Why it matters: 缺口未补前，Doramagic 不能把该能力当作可靠推荐卖点。
- Evidence: evidence.evidence_gaps | https://github.com/labsai/EDDI | Sandbox install result is 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。

### Constraint 2: 依赖 Docker 环境

- Trigger: 安装/运行入口包含 Docker 命令：docker compose up
- Host AI rule: 标注 Docker 前置条件，并提供非 Docker 路径或失败提示。
- Why it matters: 非工程用户可能没有 Docker，启动成本明显增加。
- Evidence: identity.distribution | https://github.com/labsai/EDDI | docker compose up
- 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: 安装命令尚未沙箱验证

- Trigger: 当前 install_status=documented，还只是文档/元数据线索。
- Host AI rule: 在隔离环境执行最小安装路径，记录失败原因和修复步骤。
- Why it matters: 命令可能缺步骤、过期或依赖本地环境，不能直接作为用户承诺。
- Evidence: downstream_validation.install_status | https://github.com/labsai/EDDI | install_status=documented; command=docker compose up
- 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/labsai/EDDI | 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: Quick Start 尚未实际跑通

- Trigger: quickstart_status=not_attempted。
- Host AI rule: 执行最小 repro plan，记录命令、输入样例、输出样例和失败原因。
- Why it matters: 用户只能看到安装线索，不能确信 10 分钟内能形成最小可试路径。
- Evidence: downstream_validation.quickstart_status | https://github.com/labsai/EDDI | quickstart_status=not_attempted; sandbox_quickstart_status=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。

### Constraint 6: 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/labsai/EDDI | 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/labsai/EDDI | 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: 存在安全注意事项

- Trigger: No sandbox install has been executed yet; downstream must verify before user use.
- Host AI rule: 转成明确权限清单和安全审查提示。
- Why it matters: 用户安装前需要知道权限边界和敏感操作。
- Evidence: risks.safety_notes | https://github.com/labsai/EDDI | No sandbox install has been executed yet; downstream must verify before user use.
- 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: 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/labsai/EDDI | 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 10: 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/labsai/EDDI | 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。