# printr-monorepo - 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 printr-monorepo 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_0013` supported 0.86
- **希望把专业流程带进宿主 AI 的用户**：仓库包含 Skill 文档。 证据：`packages/cli/skills/printr/SKILL.md` Claim：`clm_0014` supported 0.86

## 它能做什么

- **Token Creation (Multi-Chain)**（可做安装前预览）：Create and launch tokens across EVM chains and Solana with optional automated signing, image generation, and initial liquidity buys. 证据：`packages/mcp/src/tools/create-token.ts`, `packages/mcp/src/tools/launch-token.ts`, `packages/sdk/src/token.ts`, `packages/mcp/src/tools/quote.ts` 等 Claim：`clm_0001` supported 0.86, `clm_0004` supported 0.86
- **Transaction Signing (EVM & SVM)**（可做安装前预览）：Sign and submit blockchain transactions for EVM (via viem) and Solana (via @solana/web3.js) chains, with browser wallet support. 证据：`packages/mcp/src/tools/sign-and-submit-evm.ts`, `packages/mcp/src/tools/sign-and-submit-svm.ts`, `packages/mcp/src/tools/open-web-signer.ts`, `packages/sdk/src/evm.ts` 等 Claim：`clm_0002` supported 0.86, `clm_0003` supported 0.86, `clm_0012` supported 0.86
- **Encrypted Wallet Management**（可做安装前预览）：Generate, import, list, unlock, and remove wallets with AES-256-GCM encryption in a local keystore. 证据：`packages/mcp/src/tools/wallet.ts`, `packages/sdk/src/keystore.ts`, `packages/sdk/src/evm.ts`, `packages/mcp/src/server/wallet-sessions.ts` Claim：`clm_0002` supported 0.86, `clm_0003` supported 0.86
- **Deployment Wallet Lifecycle (Treasury Protection)**（可做安装前预览）：Create ephemeral deployment wallets funded from treasury, auto-drain after deployment, with recovery support for failed drains. 证据：`packages/mcp/src/tools/fund-deployment-wallet.ts`, `packages/mcp/src/tools/drain-deployment-wallet.ts`, `packages/mcp/src/tools/set-treasury-wallet.ts`, `packages/mcp/src/lib/drain.ts` 等 Claim：`clm_0001` supported 0.86, `clm_0004` supported 0.86
- **Balance Queries & Token Transfers**（可做安装前预览）：Query native and ERC-20/SPL token balances across chains, execute native token transfers. 证据：`packages/mcp/src/tools/get-balance.ts`, `packages/mcp/src/tools/get-token-balance.ts`, `packages/mcp/src/tools/transfer.ts`, `packages/mcp/src/tools/transfer-token.ts` 等 Claim：`clm_0005` supported 0.86
- **AI-Powered Token Image Generation**（需要安装后验证）：Generate token avatars via OpenRouter AI image models, with auto-generation fallback when creating tokens. 证据：`packages/mcp/src/tools/generate-image.ts`, `packages/sdk/src/image.ts`, `packages/mcp/src/lib/env.ts`, `packages/mcp/src/index.ts` Claim：`clm_0006` supported 0.86, `clm_0009` supported 0.86
- **Protocol Fee Management**（可做安装前预览）：Query and claim creator fees across chains, with automatic fee collection to treasury after deployments. 证据：`packages/mcp/src/tools/claim-fees.ts`, `packages/mcp/src/tools/get-creator-fees.ts`, `packages/sdk/src/fees-api.ts`, `packages/cli/skills/printr/SKILL.md` Claim：`clm_0007` supported 0.86
- **Staking Operations**（可做安装前预览）：Create stake positions on Printr token, list positions with rewards, claim rewards or withdraw unlocked principal. 证据：`packages/mcp/src/tools/create-stake-position.ts`, `packages/mcp/src/tools/get-staking-positions.ts`, `packages/mcp/src/tools/claim-staking-rewards.ts`, `packages/sdk/src/index.ts` Claim：`clm_0008` supported 0.86
- **MCP Server with Stdio Transport**（可做安装前预览）：Model Context Protocol server that exposes all tools to AI agents via stdio transport, compatible with Claude Desktop, Cursor, and other MCP clients. 证据：`packages/mcp/src/mcp.ts`, `packages/mcp/src/index.ts`, `packages/mcp/package.json`, `packages/mcp/src/lib/handler.ts` Claim：`clm_0006` supported 0.86, `clm_0009` supported 0.86, `clm_0012` supported 0.86
- **CLI for MCP Client Configuration**（可做安装前预览）：CLI tool that auto-detects AI clients and configures MCP server settings, with optional OpenRouter API key setup. 证据：`packages/cli/src/commands/setup/index.ts`, `packages/cli/src/commands/skill/index.ts`, `packages/cli/src/index.ts`, `packages/cli/package.json` Claim：`clm_0010` supported 0.86
- **Type-Safe SDK with CAIP Compliance**（可做安装前预览）：Framework-agnostic TypeScript SDK with full Zod schema validation, CAIP-2/CAIP-10 chain/address format support, and gRPC/OpenAPI client options. 证据：`packages/sdk/package.json`, `packages/sdk/src/caip.ts`, `packages/sdk/src/schemas.ts`, `packages/sdk/src/client.ts` 等 Claim：`clm_0011` supported 0.86
- **Browser Signing Server**（可做安装前预览）：Hono-based HTTP server that generates QR codes and manages transaction signing sessions for MetaMask and Phantom wallets. 证据：`packages/mcp/src/server/app.ts`, `packages/mcp/src/server/index.ts`, `packages/mcp/src/lib/qr.ts`, `packages/mcp/package.json` Claim：`clm_0002` supported 0.86, `clm_0009` supported 0.86, `clm_0012` supported 0.86

## 怎么开始

- `curl -fsSL https://bun.sh/install | bash` 证据：`INSTALL.md` Claim：`clm_0015` supported 0.86
- `npx @printr/cli setup` 证据：`INSTALL.md` Claim：`clm_0016` supported 0.86
- `curl -fsSL https://raw.githubusercontent.com/PrintrFi/printr-mcp/main/scripts/install.sh | bash` 证据：`INSTALL.md` Claim：`clm_0017` supported 0.86
- `claude mcp add --scope user printr -- bunx @printr/mcp@latest` 证据：`INSTALL.md` Claim：`clm_0018` supported 0.86
- `claude mcp add --scope user printr -- npx -y @printr/mcp@latest` 证据：`INSTALL.md` Claim：`clm_0019` supported 0.86
- `npx skills add PrintrFi/printr-mcp` 证据：`README.md` Claim：`clm_0020` supported 0.86, `clm_0021` supported 0.86
- `npx skills add PrintrFi/printr-mcp --list` 证据：`README.md` Claim：`clm_0021` supported 0.86
- `npm install -g @printr/cli` 证据：`packages/cli/README.md` Claim：`clm_0022` supported 0.86
- `npx @printr/mcp@latest` 证据：`packages/cli/README.md` Claim：`clm_0023` supported 0.86
- `npm install @printr/sdk` 证据：`packages/sdk/README.md` Claim：`clm_0024` supported 0.86

## 继续前判断卡

- **当前建议**：需要管理员/安全审批
- **为什么**：继续前可能涉及密钥、账号、外部服务或敏感上下文，建议先经过管理员或安全审批。

### 30 秒判断

- **现在怎么做**：需要管理员/安全审批
- **最小安全下一步**：先跑 Prompt Preview；若涉及凭证或企业环境，先审批再试装
- **先别相信**：工具权限边界不能在安装前相信。
- **继续会触碰**：命令执行、宿主 AI 配置、本地环境或项目文件

### 现在可以相信

- **适合人群线索：正在使用 Claude/Codex/Cursor/Gemini 等宿主 AI 的开发者**（supported）：有 supported claim 或项目证据支撑，但仍不等于真实安装效果。 证据：`README.md` Claim：`clm_0013` supported 0.86
- **适合人群线索：希望把专业流程带进宿主 AI 的用户**（supported）：有 supported claim 或项目证据支撑，但仍不等于真实安装效果。 证据：`packages/cli/skills/printr/SKILL.md` Claim：`clm_0014` supported 0.86
- **能力存在：Token Creation (Multi-Chain)**（supported）：可以相信项目包含这类能力线索；是否适合你的具体任务仍要试用或安装后验证。 证据：`packages/mcp/src/tools/create-token.ts`, `packages/mcp/src/tools/launch-token.ts`, `packages/sdk/src/token.ts`, `packages/mcp/src/tools/quote.ts` 等 Claim：`clm_0001` supported 0.86, `clm_0004` supported 0.86
- **能力存在：Transaction Signing (EVM & SVM)**（supported）：可以相信项目包含这类能力线索；是否适合你的具体任务仍要试用或安装后验证。 证据：`packages/mcp/src/tools/sign-and-submit-evm.ts`, `packages/mcp/src/tools/sign-and-submit-svm.ts`, `packages/mcp/src/tools/open-web-signer.ts`, `packages/sdk/src/evm.ts` 等 Claim：`clm_0002` supported 0.86, `clm_0003` supported 0.86, `clm_0012` supported 0.86
- **能力存在：Encrypted Wallet Management**（supported）：可以相信项目包含这类能力线索；是否适合你的具体任务仍要试用或安装后验证。 证据：`packages/mcp/src/tools/wallet.ts`, `packages/sdk/src/keystore.ts`, `packages/sdk/src/evm.ts`, `packages/mcp/src/server/wallet-sessions.ts` Claim：`clm_0002` supported 0.86, `clm_0003` supported 0.86
- **能力存在：Deployment Wallet Lifecycle (Treasury Protection)**（supported）：可以相信项目包含这类能力线索；是否适合你的具体任务仍要试用或安装后验证。 证据：`packages/mcp/src/tools/fund-deployment-wallet.ts`, `packages/mcp/src/tools/drain-deployment-wallet.ts`, `packages/mcp/src/tools/set-treasury-wallet.ts`, `packages/mcp/src/lib/drain.ts` 等 Claim：`clm_0001` supported 0.86, `clm_0004` supported 0.86

### 现在还不能相信

- **工具权限边界不能在安装前相信。**（unverified）：MCP/tool 类项目通常会触碰文件、网络、浏览器或外部 API，必须真实检查权限和日志。
- **真实输出质量不能在安装前相信。**（unverified）：Prompt Preview 只能展示引导方式，不能证明真实项目中的结果质量。
- **宿主 AI 版本兼容性不能在安装前相信。**（unverified）：Claude、Cursor、Codex、Gemini 等宿主加载规则和版本差异必须在真实环境验证。
- **不会污染现有宿主 AI 行为，不能直接相信。**（inferred）：Skill、plugin、AGENTS/CLAUDE/GEMINI 指令可能改变宿主 AI 的默认行为。 证据：`AGENTS.md`, `CLAUDE.md`, `packages/cli/skills/printr/SKILL.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。
- **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：`INSTALL.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：`INSTALL.md`, `README.md`, `packages/cli/README.md`, `packages/sdk/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：`AGENTS.md`, `CLAUDE.md`, `packages/cli/skills/printr/SKILL.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：`packages/mcp/src/index.ts`, `packages/mcp/src/lib/env.ts`, `packages/mcp/src/tools/generate-image.ts`, `packages/sdk/src/image.ts`
- **环境变量 / API Key**：项目入口文档明确出现 API key、token、secret 或账号凭证配置。 原因：如果真实安装需要凭证，应先使用测试凭证并经过权限/合规判断。 证据：`CHANGELOG.md`, `INSTALL.md`, `README.md`, `docs/consumer-integration-analysis.md` 等
- **宿主 AI 上下文**：AI Context Pack、Prompt Preview、Skill 路由、风险规则和项目事实。 原因：导入上下文会影响宿主 AI 后续判断，必须避免把未验证项包装成事实。

### 最小安全下一步

- **先跑 Prompt Preview**：用安装前交互式试用判断工作方式是否匹配，不需要授权或改环境。（适用：任何项目都适用，尤其是输出质量未知时。）
- **只在隔离目录或测试账号试装**：避免安装命令污染主力宿主 AI、真实项目或用户主目录。（适用：存在命令执行、插件配置或本地写入线索时。）
- **先备份宿主 AI 配置**：Skill、plugin、规则文件可能改变 Claude/Cursor/Codex 的默认行为。（适用：存在插件 manifest、Skill 或宿主规则入口时。）
- **不要使用真实生产凭证**：环境变量/API key 一旦进入宿主或工具链，可能产生账号和合规风险。（适用：出现 API、TOKEN、KEY、SECRET 等环境线索时。）
- **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。
- **准备撤销测试 API key 或 token**：测试凭证泄露或误用时，可以快速止损。
- **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_0026` 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：`INSTALL.md`, `README.md`, `packages/cli/README.md`, `packages/sdk/README.md` Claim：`clm_0027` supported 0.86
- **风险**： 处理方式：
- **风险**： 处理方式：
- **风险**： 处理方式：
- **风险**： 处理方式：
- **风险**： 处理方式：
- **风险**： 处理方式：
- **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

- **Token Creation (Multi-Chain)**：先基于 role_skill_index / evidence_index 帮用户挑选可用角色、Skill 或工作流。 边界：可做安装前 Prompt 体验。 证据：`packages/mcp/src/tools/create-token.ts`, `packages/mcp/src/tools/launch-token.ts`, `packages/sdk/src/token.ts`, `packages/mcp/src/tools/quote.ts` 等 Claim：`clm_0001` supported 0.86, `clm_0004` supported 0.86
- **Transaction Signing (EVM & SVM)**：先基于 role_skill_index / evidence_index 帮用户挑选可用角色、Skill 或工作流。 边界：可做安装前 Prompt 体验。 证据：`packages/mcp/src/tools/sign-and-submit-evm.ts`, `packages/mcp/src/tools/sign-and-submit-svm.ts`, `packages/mcp/src/tools/open-web-signer.ts`, `packages/sdk/src/evm.ts` 等 Claim：`clm_0002` supported 0.86, `clm_0003` supported 0.86, `clm_0012` supported 0.86
- **Encrypted Wallet Management**：先基于 role_skill_index / evidence_index 帮用户挑选可用角色、Skill 或工作流。 边界：可做安装前 Prompt 体验。 证据：`packages/mcp/src/tools/wallet.ts`, `packages/sdk/src/keystore.ts`, `packages/sdk/src/evm.ts`, `packages/mcp/src/server/wallet-sessions.ts` Claim：`clm_0002` supported 0.86, `clm_0003` supported 0.86
- **Deployment Wallet Lifecycle (Treasury Protection)**：先基于 role_skill_index / evidence_index 帮用户挑选可用角色、Skill 或工作流。 边界：可做安装前 Prompt 体验。 证据：`packages/mcp/src/tools/fund-deployment-wallet.ts`, `packages/mcp/src/tools/drain-deployment-wallet.ts`, `packages/mcp/src/tools/set-treasury-wallet.ts`, `packages/mcp/src/lib/drain.ts` 等 Claim：`clm_0001` supported 0.86, `clm_0004` supported 0.86
- **Balance Queries & Token Transfers**：先基于 role_skill_index / evidence_index 帮用户挑选可用角色、Skill 或工作流。 边界：可做安装前 Prompt 体验。 证据：`packages/mcp/src/tools/get-balance.ts`, `packages/mcp/src/tools/get-token-balance.ts`, `packages/mcp/src/tools/transfer.ts`, `packages/mcp/src/tools/transfer-token.ts` 等 Claim：`clm_0005` supported 0.86
- **AI-Powered Token Image Generation**：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：`packages/mcp/src/tools/generate-image.ts`, `packages/sdk/src/image.ts`, `packages/mcp/src/lib/env.ts`, `packages/mcp/src/index.ts` Claim：`clm_0006` supported 0.86, `clm_0009` supported 0.86
- **Protocol Fee Management**：先基于 role_skill_index / evidence_index 帮用户挑选可用角色、Skill 或工作流。 边界：可做安装前 Prompt 体验。 证据：`packages/mcp/src/tools/claim-fees.ts`, `packages/mcp/src/tools/get-creator-fees.ts`, `packages/sdk/src/fees-api.ts`, `packages/cli/skills/printr/SKILL.md` Claim：`clm_0007` supported 0.86
- **Staking Operations**：先基于 role_skill_index / evidence_index 帮用户挑选可用角色、Skill 或工作流。 边界：可做安装前 Prompt 体验。 证据：`packages/mcp/src/tools/create-stake-position.ts`, `packages/mcp/src/tools/get-staking-positions.ts`, `packages/mcp/src/tools/claim-staking-rewards.ts`, `packages/sdk/src/index.ts` Claim：`clm_0008` supported 0.86
- **MCP Server with Stdio Transport**：先基于 role_skill_index / evidence_index 帮用户挑选可用角色、Skill 或工作流。 边界：可做安装前 Prompt 体验。 证据：`packages/mcp/src/mcp.ts`, `packages/mcp/src/index.ts`, `packages/mcp/package.json`, `packages/mcp/src/lib/handler.ts` Claim：`clm_0006` supported 0.86, `clm_0009` supported 0.86, `clm_0012` supported 0.86
- **CLI for MCP Client Configuration**：先基于 role_skill_index / evidence_index 帮用户挑选可用角色、Skill 或工作流。 边界：可做安装前 Prompt 体验。 证据：`packages/cli/src/commands/setup/index.ts`, `packages/cli/src/commands/skill/index.ts`, `packages/cli/src/index.ts`, `packages/cli/package.json` Claim：`clm_0010` supported 0.86
- **Type-Safe SDK with CAIP Compliance**：先基于 role_skill_index / evidence_index 帮用户挑选可用角色、Skill 或工作流。 边界：可做安装前 Prompt 体验。 证据：`packages/sdk/package.json`, `packages/sdk/src/caip.ts`, `packages/sdk/src/schemas.ts`, `packages/sdk/src/client.ts` 等 Claim：`clm_0011` supported 0.86
- **Browser Signing Server**：先基于 role_skill_index / evidence_index 帮用户挑选可用角色、Skill 或工作流。 边界：可做安装前 Prompt 体验。 证据：`packages/mcp/src/server/app.ts`, `packages/mcp/src/server/index.ts`, `packages/mcp/src/lib/qr.ts`, `packages/mcp/package.json` Claim：`clm_0002` supported 0.86, `clm_0009` supported 0.86, `clm_0012` supported 0.86

### 上下文规模

- 文件总数：218
- 重要文件覆盖：40/218
- 证据索引条目：77
- 角色 / Skill 条目：1

### 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 printr-monorepo 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 printr-monorepo 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 printr-monorepo 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 1 个角色 / Skill / 项目文档条目。

- **printr**（skill）：Launch cross-chain tokens via Printr MCP. Use to deploy tokens, check wallet balances, manage keystore wallets, transfer native or ERC-20 / SPL tokens on Base, Ethereum, Arbitrum, Solana, other supported chains. 激活提示：当用户任务与“printr”描述的流程高度相关时，先用它做安装前体验，再决定是否安装。 证据：`packages/cli/skills/printr/SKILL.md`

## 证据索引

- 共索引 77 条证据。

- **@printr/mcp**（documentation）：! npm version https://img.shields.io/npm/v/@printr/mcp.svg https://www.npmjs.com/package/@printr/mcp ! npm downloads https://img.shields.io/npm/dm/@printr/mcp.svg https://www.npmjs.com/package/@printr/mcp ! license https://img.shields.io/npm/l/@printr/mcp.svg https://github.com/PrintrFi/printr-mcp/blob/main/LICENSE ! CI https://github.com/PrintrFi/printr-mcp/actions/workflows/ci.yml/badge.svg https://github.com/PrintrFi/printr-mcp/actions/workflows/ci.yml ! codecov https://codecov.io/gh/PrintrFi/printr-mcp/branch/main/graph/badge.svg https://codecov.io/gh/PrintrFi/printr-mcp ! skills.sh https://skills.sh/b/PrintrFi/printr-mcp https://skills.sh/PrintrFi/printr-mcp 证据：`README.md`
- **Examples**（documentation）：Runnable examples demonstrating @printr/sdk and @printr/mcp usage. 证据：`examples/README.md`
- **@printr/cli**（documentation）：CLI for Printr https://printr.money — configure MCP servers and install agent skills for AI clients. 证据：`packages/cli/README.md`
- **@printr/mcp**（documentation）：MCP server for Printr https://printr.money — lets AI agents create, discover, and track tokens across chains. 证据：`packages/mcp/README.md`
- **@printr/sdk**（documentation）：TypeScript SDK for Printr https://printr.money — create and manage tokens across EVM chains and Solana. 证据：`packages/sdk/README.md`
- **Package**（package_manifest）：{ "name": "printr-monorepo", "private": true, "type": "module", "repository": { "type": "git", "url": "https://github.com/PrintrFi/printr-mcp" }, "license": "Apache-2.0", "workspaces": "packages/ ", "examples/ " , "scripts": { "build": "bun run build:sdk && bun run build:packages", "build:sdk": "bun run --cwd packages/sdk build", "build:packages": "bun run --filter '@printr/mcp' --filter '@printr/cli' build", "dev": "bun run --cwd packages/cli dev", "dev:mcp": "bun run --cwd packages/mcp dev", "start": "bun run --cwd packages/cli start", "start:mcp": "bun run --cwd packages/mcp start", "inspect": "bun run --cwd packages/mcp inspect", "test": "bun run --filter ' ' test", "test:all": "bun run… 证据：`package.json`
- **Package**（package_manifest）：{ "name": "example-mcp-client", "version": "0.0.0", "private": true, "type": "module", "scripts": { "start": "bun run index.ts", "test": "bun run index.ts" }, "dependencies": { "@printr/mcp": "workspace: ", "@modelcontextprotocol/sdk": "^1.29.0" } } 证据：`examples/mcp-client/package.json`
- **Package**（package_manifest）：{ "name": "example-sdk-basic", "version": "0.0.0", "private": true, "type": "module", "scripts": { "start": "bun run index.ts", "test": "bun run index.ts" }, "dependencies": { "@printr/sdk": "workspace: " } } 证据：`examples/sdk-basic/package.json`
- **Package**（package_manifest）：{ "name": "@printr/cli", "version": "0.3.2", "description": "CLI for Printr - configure MCP servers and install agent skills", "type": "module", "main": "./dist/index.js", "types": "./dist/index.d.ts", "bin": { "printr": "./dist/index.js" }, "publishConfig": { "access": "public" }, "files": "dist", "LICENSE", "README.md" , "repository": { "type": "git", "url": "https://github.com/PrintrFi/printr-mcp", "directory": "packages/cli" }, "license": "Apache-2.0", "keywords": "printr", "cli", "mcp", "ai", "agent", "skills", "solana", "evm", "token", "web3" , "engines": { "node": " =18" }, "scripts": { "build": "bun run scripts/build.ts", "dev": "bun run scripts/build.ts --watch", "start": "bun dist… 证据：`packages/cli/package.json`
- **Package**（package_manifest）：{ "name": "@printr/mcp", "version": "0.16.1", "description": "MCP server library for the Printr API - enables AI agents to launch tokens across EVM chains and Solana", "mcpName": "io.github.PrintrFi/printr", "type": "module", "main": "./dist/index.js", "types": "./dist/index.d.ts", "bin": { "printr-mcp": "./dist/index.js" }, "publishConfig": { "access": "public" }, "files": "dist", "certs", "LICENSE", "README.md" , "repository": { "type": "git", "url": "https://github.com/PrintrFi/printr-mcp", "directory": "packages/mcp" }, "license": "Apache-2.0", "keywords": "printr", "mcp", "model-context-protocol", "ai", "agent", "solana", "evm", "token", "web3", "crypto" , "engines": { "node": " =18" }… 证据：`packages/mcp/package.json`
- **Package**（package_manifest）：{ "name": "@printr/sdk", "version": "0.6.1", "description": "TypeScript SDK for the Printr API - create and manage tokens across EVM chains and Solana", "type": "module", "main": "./dist/index.js", "types": "./dist/index.d.ts", "exports": { ".": { "types": "./dist/index.d.ts", "import": "./dist/index.js" }, "./client": { "types": "./dist/client.d.ts", "import": "./dist/client.js" }, "./chains": { "types": "./dist/chains.d.ts", "import": "./dist/chains.js" }, "./evm": { "types": "./dist/evm.d.ts", "import": "./dist/evm.js" }, "./svm": { "types": "./dist/svm.d.ts", "import": "./dist/svm.js" }, "./schemas": { "types": "./dist/schemas.d.ts", "import": "./dist/schemas.js" }, "./caip": { "types":… 证据：`packages/sdk/package.json`
- **Printr MCP**（skill_instruction）：printr launch token create + sign in one call: - Omit private key → browser signer open - Omit image / image path → auto-gen need OPENROUTER API KEY 证据：`packages/cli/skills/printr/SKILL.md`
- **Printr Monorepo**（documentation）：Monorepo containing the Printr SDK and MCP server for AI agents to launch tokens across EVM chains and Solana. 证据：`AGENTS.md`
- **Claude**（documentation）：Read and follow the project conventions in AGENTS.md. 证据：`CLAUDE.md`
- **Printr MCP — Setup Instructions**（documentation）：These instructions are written for an autonomous agent. Read them fully, then execute each step in order. 证据：`INSTALL.md`
- **License**（source_file）：Apache License Version 2.0, January 2004 http://www.apache.org/licenses/ 证据：`LICENSE`
- **Consumer Integration Analysis: printr-mcp ↔ memeprintr**（documentation）：Consumer Integration Analysis: printr-mcp ↔ memeprintr 证据：`docs/consumer-integration-analysis.md`
- **Deployment Wallet Recovery Design**（documentation）：Safe, recoverable deployment wallet lifecycle for printr-mcp and memeprintr. 证据：`docs/deployment-wallet-recovery-design.md`
- **Index**（source_file）：import { Client } from "@modelcontextprotocol/sdk/client/index.js"; import { StdioClientTransport } from "@modelcontextprotocol/sdk/client/stdio.js"; ⋮---- async function main 证据：`examples/mcp-client/index.ts`
- **Index**（source_file）：import { createPrintrClient, unwrapResult, CHAIN META, env, } from "@printr/sdk"; ⋮---- async function main 证据：`examples/sdk-basic/index.ts`
- **Banner**（source_file）：import { Box, Text } from "ink"; ⋮---- export function Banner 证据：`packages/cli/src/commands/setup/components/banner.tsx`
- **Summary**（source_file）：import { Box, Text } from "ink"; import { CLIENTS } from "../lib/clients.js"; 证据：`packages/cli/src/commands/setup/components/summary.tsx`
- **Index**（source_file）：import { render } from "ink"; import { createElement } from "react"; import { SetupApp } from "./app.js"; import { parseSetupArgs } from "./lib/args.js"; ⋮---- export async function runSetup args: string : Promise 证据：`packages/cli/src/commands/setup/index.ts`
- **Banner**（source_file）：import { Box, Text } from "ink"; ⋮---- export function Banner 证据：`packages/cli/src/commands/skill/components/banner.tsx`
- **Summary**（source_file）：import { Box, Text } from "ink"; ⋮---- export function Summary 证据：`packages/cli/src/commands/skill/components/summary.tsx`
- **Index**（source_file）：import { render } from "ink"; import { createElement } from "react"; import { SkillApp } from "./app.js"; import { ALL AGENT IDS } from "./lib/agents.js"; ⋮---- function parseSkillArgs args: string : string null ⋮---- export async function runSkillInstall args: string : Promise 证据：`packages/cli/src/commands/skill/index.ts`
- **Index**（source_file）：import { version } from "../package.json"; 证据：`packages/cli/src/index.ts`
- **Env**（source_file）：import { ALCHEMY RPC TEMPLATES, rpcUrlsSchema, env as sdkEnv } from "@printr/sdk"; import { z } from "zod"; ⋮---- export type Env = z.infer ; 证据：`packages/mcp/src/lib/env.ts`
- **Logging**（source_file）：import { logger } from "@printr/sdk"; ⋮---- export function logToolExecution toolName: string, handler: input: TInput = TOutput Promise , : input: TInput = Promise 证据：`packages/mcp/src/lib/logging.ts`
- **Wallet Elicit**（source_file）：import type { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { type ChainType, chainTypeFromCaip2, checkEvmBalance, checkSvmBalance, getChainMeta, normalisePrivateKey, parseEvmCaip10, } from "@printr/sdk"; import { Keypair } from "@solana/web3.js"; import bs58 from "bs58"; import { match } from "ts-pattern"; import { privateKeyToAccount } from "viem/accounts"; import { env } from "~/lib/env.js"; import { type ActiveWallet, activeWallets } from "~/server/wallet-sessions.js"; ⋮---- export type TxContext = { type: "evm"; caip10To: string; gasLimit: number; rpcUrl?: string undefined } { type: "svm"; rpcUrl?: string undefined }; ⋮---- export type WalletResolution = { kind: "r… 证据：`packages/mcp/src/lib/wallet-elicit.ts`
- **Mcp**（source_file）：import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { createPrintrClient } from "@printr/sdk"; import { env } from "~/lib/env.js"; import { registerClaimFeesTool } from "~/tools/claim-fees.js"; import { registerClaimStakingRewardsTool } from "~/tools/claim-staking-rewards.js"; import { registerCreateStakePositionTool } from "~/tools/create-stake-position.js"; import { registerCreateTokenTool } from "~/tools/create-token.js"; import { registerDrainDeploymentWalletTool } from "~/tools/drain-deployment-wallet.js"; import { registerFundDeploymentWalletTool } from "~/tools/fund-deployment-wa… 证据：`packages/mcp/src/mcp.ts`
- **App**（source_file）：import { logger } from "@printr/sdk"; import { Hono } from "hono"; import { bodyLimit } from "hono/body-limit"; import { cors } from "hono/cors"; import { type CreateSessionInput, createSession, sessions, setResult, type TxResult, } from "./sessions.js"; ⋮---- export function buildApp 证据：`packages/mcp/src/server/app.ts`
- **Index**（source_file）：import { existsSync, readFileSync } from "node:fs"; import { createServer as createHttpsServer } from "node:https"; import { createServer as createNetServer } from "node:net"; import { dirname, join } from "node:path"; import { fileURLToPath } from "node:url"; import { serve } from "@hono/node-server"; import { buildApp } from "./app.js"; ⋮---- function loadCerts : ⋮---- function isPortFree port: number : Promise ⋮---- async function findFreePort start: number, end: number : Promise ⋮---- export async function startSessionServer : Promise 证据：`packages/mcp/src/server/index.ts`
- **Sessions**（source_file）：import { randomUUID } from "node:crypto"; ⋮---- export type ChainType = "evm" "svm"; ⋮---- export type TokenMeta = { name: string; symbol: string; description?: string undefined; image url?: string undefined; }; ⋮---- export type TxResult = { status: "success" "failed"; tx hash?: string; signature?: string; error?: string; image data?: string; }; ⋮---- export type TxSession = { token: string; chain type: ChainType; payload: unknown; token id: string; token meta?: TokenMeta undefined; rpc url?: string undefined; created at: number; expires at: number; result?: TxResult; }; ⋮---- export type CreateSessionInput = Omit ; ⋮---- export function createSession input: CreateSessionInput : TxSession… 证据：`packages/mcp/src/server/sessions.ts`
- **Wallet Sessions**（source_file）：import type { ChainType } from "@printr/sdk"; ⋮---- export type ActiveWallet = { privateKey: string; address: string; }; 证据：`packages/mcp/src/server/wallet-sessions.ts`
- **Claim Fees**（source_file）：import type { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { type ChainProtocolFeesSimple, chainTypeFromCaip2, decryptKey, type EvmPayload, formatEvmSubmitError, formatSvmSubmitError, getChainMeta, getProtocolFees, getSvmRpcUrl, listWallets, logger, type PayloadEVM, type PayloadSolana, type SvmPayload, signAndSubmitEvm, signAndSubmitSvm, transferSvm, } from "@printr/sdk"; import { err, errAsync, ok, okAsync, type Result, ResultAsync } from "neverthrow"; import { match } from "ts-pattern"; import { z } from "zod"; import { drainSvm } from "~/lib/drain.js"; import { env } from "~/lib/env.js"; import { type HandlerCtx, handler } from "~/lib/handler.js"; import { getTreasu… 证据：`packages/mcp/src/tools/claim-fees.ts`
- **Claim Staking Rewards**（source_file）：import type { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { type ChainType, type ClaimStakingRewardsResult, chainTypeFromCaip2, claimStakingRewards, type EvmPayload, formatEvmSubmitError, formatSvmSubmitError, getSvmRpcUrl, parseStakingCaip10, type SimpleTxPayload, type SvmPayload, signAndSubmitEvm, signAndSubmitSvm, toToolResponseAsync, } from "@printr/sdk"; import { err, errAsync, ok, okAsync, Result, ResultAsync } from "neverthrow"; import { z } from "zod"; import { logToolExecution } from "~/lib/logging.js"; import { getTreasuryAddress, getTreasuryKey } from "~/lib/treasury.js"; ⋮---- type ClaimOutput = z.infer ; type ClaimError = { message: string }; type CaipAcc… 证据：`packages/mcp/src/tools/claim-staking-rewards.ts`
- **Create Token**（source_file）：import type { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { buildToken, caip2ChainId, caip10Address, externalLinks, graduationThreshold, initialBuy, type PrintrClient, quoteOutput, toToolResponseAsync, } from "@printr/sdk"; import { z } from "zod"; import { logToolExecution } from "~/lib/logging.js"; ⋮---- export function registerCreateTokenTool server: McpServer, client: PrintrClient 证据：`packages/mcp/src/tools/create-token.ts`
- **Drain Deployment Wallet**（source_file）：import type { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { type ChainType, chainTypeFromCaip2, decryptKey, getActiveWalletId, getChainMeta, getEvmConfig, getLastDeploymentWalletId, getWallet, } from "@printr/sdk"; import { err, errAsync, ok, type Result, type ResultAsync } from "neverthrow"; import { match } from "ts-pattern"; import { z } from "zod"; import { type DrainError, type DrainResult, drainEvm, drainSvm, type ResolvedWallet, } from "~/lib/drain.js"; import { env } from "~/lib/env.js"; import { type HandlerCtx, handler } from "~/lib/handler.js"; import { getTreasuryKeyOrError } from "~/lib/treasury.js"; import { activeWallets } from "~/server/wallet-sessions… 证据：`packages/mcp/src/tools/drain-deployment-wallet.ts`
- **Fund Deployment Wallet**（source_file）：import { randomUUID } from "node:crypto"; import { accessSync, constants } from "node:fs"; import { dirname } from "node:path"; import type { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { addWallet, type ChainType, chainTypeFromCaip2, encryptKey, executeTransfer, getChainMeta, keystorePath, logger, normalisePrivateKey, parseCaip2, setActiveWalletId, setLastDeploymentWalletId, } from "@printr/sdk"; import { Keypair } from "@solana/web3.js"; import bs58 from "bs58"; import { err, errAsync, ok, okAsync, type Result, type ResultAsync } from "neverthrow"; import { match } from "ts-pattern"; import { generatePrivateKey, privateKeyToAccount } from "viem/accounts"; import { z… 证据：`packages/mcp/src/tools/fund-deployment-wallet.ts`
- **Generate Image**（source_file）：import { writeFile } from "node:fs/promises"; import { tmpdir } from "node:os"; import { join } from "node:path"; import type { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { generateImageFromPrompt } from "@printr/sdk"; import { ResultAsync } from "neverthrow"; import { z } from "zod"; import { env } from "~/lib/env.js"; import { logToolExecution } from "~/lib/logging.js"; ⋮---- export function registerGenerateImageTool server: McpServer, openrouterApiKey: string : void 证据：`packages/mcp/src/tools/generate-image.ts`
- **Get Balance**（source_file）：import type { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { CHAIN META, fetchNativeBalance, getChainMeta, isSupportedNamespace, parseCaip10, type SimpleBalanceResult, toCaip2, toolError, toolOk, } from "@printr/sdk"; import { z } from "zod"; import { logToolExecution } from "~/lib/logging.js"; ⋮---- type BalanceOutput = z.infer ; ⋮---- const buildOutput = account: string, caip2: string, chainName: string, result: SimpleBalanceResult, : BalanceOutput = ⋮---- export function registerGetBalanceTool server: McpServer : void 证据：`packages/mcp/src/tools/get-balance.ts`
- **Get Deployments**（source_file）：import type { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { type PrintrClient, tokenId, toToolResponseAsync, unwrapResultAsync } from "@printr/sdk"; import { z } from "zod"; import { logToolExecution } from "~/lib/logging.js"; ⋮---- export function registerGetDeploymentsTool server: McpServer, client: PrintrClient 证据：`packages/mcp/src/tools/get-deployments.ts`
- **Get Staking Positions**（source_file）：import type { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { formatStakingCaip10, listStakePositionsWithRewards, parseStakingCaip10, type SimpleStakePositionWithRewards, toToolResponseAsync, } from "@printr/sdk"; import { ResultAsync } from "neverthrow"; import { z } from "zod"; import { logToolExecution } from "~/lib/logging.js"; import { getTreasuryAddress } from "~/lib/treasury.js"; ⋮---- type CaipAccount = { chainId: string; address: string }; type OutputPosition = z.infer ; ⋮---- function defaultOwnersFromTreasury : CaipAccount ⋮---- function hasPositiveAtomic amount: ⋮---- function toOutputPosition pos: SimpleStakePositionWithRewards, now: Date, : OutputPosition… 证据：`packages/mcp/src/tools/get-staking-positions.ts`
- **Get Token Balance**（source_file）：import type { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { CHAIN META, fetchTokenBalance, getChainMeta, isSupportedNamespace, parseCaip10, type SimpleBalanceResult, toCaip2, toolError, toolOk, } from "@printr/sdk"; import { z } from "zod"; import { logToolExecution } from "~/lib/logging.js"; ⋮---- type TokenBalanceOutput = z.infer ; ⋮---- const buildOutput = token: string, wallet: string, chain: string, chainName: string, result: SimpleBalanceResult, : TokenBalanceOutput = ⋮---- export function registerGetTokenBalanceTool server: McpServer : void 证据：`packages/mcp/src/tools/get-token-balance.ts`
- **Get Token**（source_file）：import type { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { type PrintrClient, tokenId, toToolResponseAsync, unwrapResultAsync } from "@printr/sdk"; import { z } from "zod"; import { logToolExecution } from "~/lib/logging.js"; ⋮---- export function registerGetTokenTool server: McpServer, client: PrintrClient 证据：`packages/mcp/src/tools/get-token.ts`
- **Launch Token**（source_file）：import type { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { buildToken, type ChainType, caip2ChainId, caip10Address, chainTypeFromCaip2, type EvmPayload, externalLinks, formatEvmSubmitError, formatSvmSubmitError, getActiveWalletId, getChainMeta, getEvmConfig, graduationThreshold, initialBuy, logger, PrintrApiError, type PrintrClient, quoteOutput, type SvmPayload, signAndSubmitEvm, signAndSubmitSvm, type ToolResponse, toToolResponseAsync, } from "@printr/sdk"; import { ResultAsync } from "neverthrow"; import { match } from "ts-pattern"; import { z } from "zod"; import { drainEvm, drainSvm, type ResolvedWallet } from "~/lib/drain.js"; import { env } from "~/lib/env.js";… 证据：`packages/mcp/src/tools/launch-token.ts`
- **Open Web Signer**（source_file）：import type { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { z } from "zod"; import { env } from "~/lib/env.js"; import { logToolExecution } from "~/lib/logging.js"; import { appendQr } from "~/lib/qr.js"; import { createSession, LOCAL SESSION ORIGIN, startSessionServer } from "~/server"; ⋮---- export function registerOpenWebSignerTool server: McpServer : void 证据：`packages/mcp/src/tools/open-web-signer.ts`
- **Quote**（source_file）：import type { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { caip2ChainId, graduationThreshold, initialBuy, type PrintrClient, type paths, quoteOutput, toToolResponseAsync, unwrapResultAsync, } from "@printr/sdk"; import { z } from "zod"; import { logToolExecution } from "~/lib/logging.js"; ⋮---- type QuoteRequestBody = paths "/print/quote" "post" "requestBody" "content" "application/json" ; ⋮---- export function registerQuoteTool server: McpServer, client: PrintrClient 证据：`packages/mcp/src/tools/quote.ts`
- **Sign And Submit Evm**（source_file）：import type { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { caip10ToChainId, formatEvmSubmitError, signAndSubmitEvm, toolError, toolOk, } from "@printr/sdk"; import { z } from "zod"; import { env } from "~/lib/env.js"; import { logToolExecution } from "~/lib/logging.js"; import { insufficientFundsMessage, resolveWallet } from "~/lib/wallet-elicit.js"; ⋮---- export function registerSignAndSubmitEvmTool server: McpServer : void ⋮---- const submit = async key: string 证据：`packages/mcp/src/tools/sign-and-submit-evm.ts`
- **Sign And Submit Svm**（source_file）：import type { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { caip10ToChainId, formatSvmSubmitError, signAndSubmitSvm, toolError, toolOk, } from "@printr/sdk"; import { z } from "zod"; import { env } from "~/lib/env.js"; import { logToolExecution } from "~/lib/logging.js"; import { insufficientFundsMessage, resolveWallet } from "~/lib/wallet-elicit.js"; ⋮---- export function registerSignAndSubmitSvmTool server: McpServer : void ⋮---- const submit = async key: string 证据：`packages/mcp/src/tools/sign-and-submit-svm.ts`
- **Transfer Token**（source_file）：import type { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { CHAIN META, type ChainMeta, executeTokenTransfer, getChainMeta, isSupportedNamespace, namespaceToChainType, parseCaip10, toCaip2, toToolResponseAsync, } from "@printr/sdk"; import { err, ok, type Result } from "neverthrow"; import { match } from "ts-pattern"; import { z } from "zod"; import { logToolExecution } from "~/lib/logging.js"; import { activeWallets } from "~/server/wallet-sessions.js"; ⋮---- type TransferToolError = { message: string }; ⋮---- const getPrivateKey = namespace: string, providedKey?: string : string null = ⋮---- type ParsedInput = { namespace: string; chainRef: string; address: string;… 证据：`packages/mcp/src/tools/transfer-token.ts`
- **Transfer**（source_file）：import type { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { CHAIN META, type ChainMeta, executeTransfer, getChainMeta, isSupportedNamespace, namespaceToChainType, parseCaip10, toCaip2, toToolResponseAsync, } from "@printr/sdk"; import { err, ok, type Result } from "neverthrow"; import { z } from "zod"; import { logToolExecution } from "~/lib/logging.js"; import { activeWallets } from "~/server/wallet-sessions.js"; ⋮---- type TransferToolError = { message: string }; ⋮---- const getPrivateKey = namespace: string, providedKey?: string : string null = ⋮---- type ParsedInput = { namespace: string; chainRef: string; address: string; caip2: string; meta: ChainMeta; key: stri… 证据：`packages/mcp/src/tools/transfer.ts`
- **Balance Lite**（source_file）：import { err, errAsync, ok, ResultAsync } from "neverthrow"; import { z } from "zod"; import type { ChainMeta } from "./chains.js"; import { getRpcUrl, toCaip2 } from "./chains.js"; ⋮---- export type SimpleBalanceResult = { readonly balance atomic: string; readonly balance formatted: string; readonly symbol: string; readonly decimals: number; }; ⋮---- export type BalanceError = "no rpc" "fetch failed" "chain mismatch"; ⋮---- function rpcCall rpcUrl: string, method: string, params: readonly unknown ⋮---- export function formatUnits value: bigint, decimals: number : string ⋮---- function stripHexPrefix hex: string : string ⋮---- function pad32 addressHex: string : string ⋮---- function encode… 证据：`packages/sdk/src/balance-lite.ts`
- **Balance**（source_file）：import { Connection, PublicKey } from "@solana/web3.js"; import { err, errAsync, ok, type Result, ResultAsync } from "neverthrow"; import { createPublicClient, erc20Abi, formatUnits, http } from "viem"; import type { ChainMeta } from "./chains.js"; import { createViemChain, getChainMeta, getRpcUrl, toCaip2 } from "./chains.js"; import { PublicContractClient } from "./public-contract-client.js"; import { getSvmRpcUrl } from "./svm.js"; ⋮---- export type SimpleBalanceResult = { readonly balance atomic: string; readonly balance formatted: string; readonly symbol: string; readonly decimals: number; }; ⋮---- export type BalanceInfo = { address: string; balance: bigint; balanceFormatted: string;… 证据：`packages/sdk/src/balance.ts`
- **Caip**（source_file）：export type ChainType = "evm" "svm"; ⋮---- export type ParsedCaip2 = { readonly namespace: string; readonly chainRef: string; }; ⋮---- export type ParsedCaip10 = ParsedCaip2 & { readonly address: string; }; ⋮---- export type SupportedNamespace = "eip155" "solana"; ⋮---- export function parseCaip2 caip2: string : ParsedCaip2 null ⋮---- export function parseCaip10 caip10: string : ParsedCaip10 null ⋮---- export const toCaip2 = ⋮---- export const isSupportedNamespace = ns: string : ns is SupportedNamespace ⋮---- export const namespaceToChainType = namespace: string : ChainType ⋮---- export const chainTypeFromCaip2 = caip2: string : ChainType 证据：`packages/sdk/src/caip.ts`
- **Client**（source_file）：import { err, ok, type Result, ResultAsync } from "neverthrow"; import createClient from "openapi-fetch"; ⋮---- import type { paths } from "./api.gen.js"; ⋮---- export type PrintrClient = ReturnType ; ⋮---- export interface ClientConfig { apiKey: string; baseUrl: string; } ⋮---- export class PrintrApiError extends Error ⋮---- constructor public readonly status: number, public readonly detail: string, ⋮---- export function createPrintrClient config: ClientConfig ⋮---- export function unwrapResult result: { data?: T; error?: unknown; response: Response; } : Result ⋮---- function extractErrorDetail error: unknown, response: Response : string ⋮---- export function toToolResponse result: Result… 证据：`packages/sdk/src/client.ts`
- **Env**（source_file）：import { z } from "zod"; ⋮---- export type Env = z.infer ; 证据：`packages/sdk/src/env.ts`
- **Evm**（source_file）：import { err, errAsync, ok, okAsync, type Result, ResultAsync } from "neverthrow"; import { type Address, createPublicClient, createWalletClient, type Hex, http } from "viem"; import { privateKeyToAccount } from "viem/accounts"; import { createViemChain, getChainMeta, getRpcUrls } from "./chains.js"; import { ensureHex } from "./hex.js"; import { type RpcInput, withRpcFallback } from "./rpc.js"; ⋮---- const toMessage = e: unknown : string ⋮---- export type ParseEvmCaip10Error = { kind: "malformed"; input: string } { kind: "invalid chain id"; input: string }; ⋮---- export function tryParseEvmCaip10 caip10: string, : Result ⋮---- const broadcast = rpc: string : Promise ⋮---- const waitReceipt… 证据：`packages/sdk/src/evm.ts`
- **Image**（source_file）：import { OpenRouter } from "@openrouter/sdk"; import { err, errAsync, ok, okAsync, ResultAsync } from "neverthrow"; import type SharpDefault from "sharp"; import { env } from "./env.js"; ⋮---- type SharpFactory = typeof SharpDefault; type FsPromisesModule = typeof import "node:fs/promises" ; type PathModule = typeof import "node:path" ; ⋮---- export type ImageError = { message: string }; ⋮---- function loadSharp : ResultAsync ⋮---- function loadFsPromises : ResultAsync ⋮---- function loadPath : ResultAsync ⋮---- export function buildImagePrompt userPrompt: string : string ⋮---- function buildTokenImagePrompt name: string, symbol: string, description: string : string ⋮---- export interface G… 证据：`packages/sdk/src/image.ts`
- 其余 17 条证据见 `AI_CONTEXT_PACK.json` 或 `EVIDENCE_INDEX.json`。

## 宿主 AI 必须遵守的规则

- **把本资产当作开工前上下文，而不是运行环境。**：AI Context Pack 只包含证据化项目理解，不包含目标项目的可执行状态。 证据：`README.md`, `examples/README.md`, `packages/cli/README.md`
- **回答用户时区分可预览内容与必须安装后才能验证的内容。**：安装前体验的消费者价值来自降低误装和误判，而不是伪装成真实运行。 证据：`README.md`, `examples/README.md`, `packages/cli/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: README.md, package.json, packages/mcp/package.json, packages/sdk/package.json, packages/cli/package.json
- **安装配置**：importance `high`
  - source_paths: README.md, packages/mcp/README.md, packages/cli/README.md, packages/mcp/src/lib/env.ts, INSTALL.md
- **Monorepo 结构**：importance `high`
  - source_paths: package.json, packages/sdk/package.json, packages/mcp/package.json, packages/cli/package.json, tsconfig.base.json
- **SDK 架构**：importance `high`
  - source_paths: packages/sdk/src/index.ts, packages/sdk/src/client.ts, packages/sdk/src/token.ts, packages/sdk/src/balance.ts, packages/sdk/src/balance-lite.ts
- **MCP 服务器架构**：importance `high`
  - source_paths: packages/mcp/src/index.ts, packages/mcp/src/mcp.ts, packages/mcp/src/server/index.ts, packages/mcp/src/server/app.ts, packages/mcp/src/server/sessions.ts
- **MCP 工具列表**：importance `high`
  - source_paths: packages/mcp/src/tools/quote.ts, packages/mcp/src/tools/create-token.ts, packages/mcp/src/tools/launch-token.ts, packages/mcp/src/tools/get-token.ts, packages/mcp/src/tools/get-deployments.ts
- **钱包与签名**：importance `high`
  - source_paths: packages/mcp/src/tools/open-web-signer.ts, packages/mcp/src/tools/sign-and-submit-evm.ts, packages/mcp/src/tools/sign-and-submit-svm.ts, packages/mcp/src/server/wallet-sessions.ts, packages/mcp/src/lib/wallet-elicit.ts
- **交易工具**：importance `medium`
  - source_paths: packages/mcp/src/tools/quote.ts, packages/mcp/src/tools/create-stake-position.ts, packages/mcp/src/tools/claim-staking-rewards.ts, packages/mcp/src/tools/get-staking-positions.ts, packages/mcp/src/tools/claim-fees.ts

## Repo Inspection Evidence / 源码检查证据

- repo_clone_verified: true
- repo_inspection_verified: true
- repo_commit: `0b7d4a7ca840049e893abacbeabdf3c58e17850d`
- inspected_files: `package.json`, `README.md`, `docs/deployment-wallet-recovery-design.md`, `docs/consumer-integration-analysis.md`, `examples/README.md`, `examples/sdk-basic/index.ts`, `examples/sdk-basic/package.json`, `examples/mcp-client/index.ts`, `examples/mcp-client/package.json`, `packages/cli/package.json`, `packages/cli/README.md`, `packages/cli/tsconfig.json`, `packages/cli/CHANGELOG.md`, `packages/mcp/package.json`, `packages/mcp/README.md`, `packages/mcp/tsconfig.json`, `packages/mcp/CHANGELOG.md`, `packages/sdk/package.json`, `packages/sdk/README.md`, `packages/sdk/tsconfig.json`

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：Epic A: SDK Workers-ready

- 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：Epic A: SDK Workers-ready
- Host AI rule: 来源问题仍为 open，Pack Agent 需要复核是否仍影响当前版本。
- Why it matters: 可能增加新用户试用和生产接入成本。
- Evidence: community_evidence:github | cevd_29b9aed5d33e473983c9b6d2f328e9a3 | https://github.com/PrintrFi/printr-mcp/issues/103 | 来源讨论提到 node 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 2: 仓库名和安装名不一致

- Trigger: 仓库名 `printr-mcp` 与安装入口 `@printr/mcp` 不完全一致。
- Host AI rule: 在 npm/PyPI/GitHub 上确认包名映射和官方 README 说明。
- Why it matters: 用户照着仓库名搜索包或照着包名找仓库时容易走错入口。
- Evidence: identity.distribution | github_repo:1160348832 | https://github.com/PrintrFi/printr-mcp | repo=printr-mcp; install=@printr/mcp
- 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: 来源证据：feat(sdk): extract signer primitive to @printr/sdk/signer

- 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：feat(sdk): extract signer primitive to @printr/sdk/signer
- Host AI rule: 来源问题仍为 open，Pack Agent 需要复核是否仍影响当前版本。
- Why it matters: 可能阻塞安装或首次运行。
- Evidence: community_evidence:github | cevd_5d0a8de4311d400b8906158d86986f93 | https://github.com/PrintrFi/printr-mcp/issues/100 | 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 4: 来源证据：fix(sdk): emit Workers-friendly bundle, drop createRequire shim

- 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：fix(sdk): emit Workers-friendly bundle, drop createRequire shim
- 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: 可能阻塞安装或首次运行。
- Evidence: community_evidence:github | cevd_f67e8511c09a4830815ffe723906fb6e | https://github.com/PrintrFi/printr-mcp/issues/97 | 来源讨论提到 node 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 5: 来源证据：Environment discriminant + memoized getConfigs(env)

- Trigger: GitHub 社区证据显示该项目存在一个配置相关的待验证问题：Environment discriminant + memoized getConfigs(env)
- Host AI rule: 来源问题仍为 open，Pack Agent 需要复核是否仍影响当前版本。
- Why it matters: 可能增加新用户试用和生产接入成本。
- Evidence: community_evidence:github | cevd_559212dc171940c4b45b5e0a598cefb2 | https://github.com/PrintrFi/printr-mcp/issues/82 | 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 6: 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 | github_repo:1160348832 | https://github.com/PrintrFi/printr-mcp | 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 7: 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 | github_repo:1160348832 | https://github.com/PrintrFi/printr-mcp | 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。

### Constraint 8: 下游验证发现风险项

- Trigger: no_demo
- Host AI rule: 进入安全/权限治理复核队列。
- Why it matters: 下游已经要求复核，不能在页面中弱化。
- Evidence: downstream_validation.risk_items | github_repo:1160348832 | https://github.com/PrintrFi/printr-mcp | 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 9: 存在评分风险

- Trigger: no_demo
- Host AI rule: 把风险写入边界卡，并确认是否需要人工复核。
- Why it matters: 风险会影响是否适合普通用户安装。
- Evidence: risks.scoring_risks | github_repo:1160348832 | https://github.com/PrintrFi/printr-mcp | 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: 来源证据：ABI codegen pipeline for printr contracts

- Trigger: GitHub 社区证据显示该项目存在一个安全/权限相关的待验证问题：ABI codegen pipeline for printr contracts
- Host AI rule: 来源问题仍为 open，Pack Agent 需要复核是否仍影响当前版本。
- Why it matters: 可能影响授权、密钥配置或安全边界。
- Evidence: community_evidence:github | cevd_1a90df3c69554032b209fe1f2740be23 | https://github.com/PrintrFi/printr-mcp/issues/85 | 来源讨论提到 npm 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。