Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/flasherses/claude-code-dotfiles/project-engineering-initnpx skills add flasherses/claude-code-dotfiles --skill project-engineering-initgit clone --depth 1 https://github.com/flasherses/claude-code-dotfilesWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00103 | $0.04786 |
| Opus 5 | $0.00051 | $0.02393 |
| Sonnet 5 | $0.00021 | $0.00957 |
| Haiku 4.5 | $0.00010 | $0.00479 |
Grade D, and why
project-engineering-init scanned grade D with 3 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
"Edit(<项目根绝对路径>/.claude/settings.json)" Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
"Bash(rm -rf*)", "Bash(curl*|*sh*)", Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
"Bash(rm -rf*)", "Bash(curl*|*sh*)", How it starts
The opening of the file, as written. The whole thing — 419 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Project Engineering Init
Quick Reference
1. Explore → 读构建文件 + 入口代码 → 输出 10 项探测结果
2. Plan → 对照决策树 → 列出生成清单 → 用户确认
3. Generate → 逐文件生成：CLAUDE.md → .claudeignore → settings.json → rules → skills/commands/agents/hooks
4. Validate → 文件存在 + JSON 格式 + 脚本语法 → 报告
何时不用: 只改一个配置文件、已有完整工程体系只需更新、非项目根目录。
目标
为任意项目生成完整的 Claude Code 项目级工程化体系。一次执行，覆盖基础层 + 执行层 + 治理层所有项目级配置。
适用判断
执行前:
# 确认在项目根目录
Test-Path ".\pom.xml" # 或 package.json / go.mod / Cargo.toml / requirements.txt
跳过此 Skill 如果：
- 用户只要求修改单个已有配置文件
- 项目已有完整的
.claude/目录且用户只想微调 - 当前目录不是项目根目录（无构建文件）
阶段 1: Explore — 探测项目
读取以下文件收集 10 项信息：
- 构建文件 (
pom.xml/package.json/go.mod/Cargo.toml/requirements.txt) — 取前 60 行 - 入口文件 (
*Application.java/main.py/app.js/index.ts/main.go) — 取前 30 行 - 目录结构 — 顶层 2 层
必须输出的 10 项:
| # | 信息 | 来源 |
|---|---|---|
| 1 | 项目类型 | 构建文件名 |
| 2 | 语言与版本 | 构建文件中的 version 字段 |
| 3 | 框架 | 入口文件的 import/package |
| 4 | 构建工具 | 构建文件名 + scripts |
| 5 | 数据库 | 构建文件依赖 + 配置文件 |
| 6 | 多模块? | 父 POM 的 <modules> 或 monorepo 结构 |
| 7 | 有测试? | src/test/ / tests/ / __tests__/ / spec/ 是否存在 |
| 8 | 有敏感配置? | .env / application-*.yml / secrets.* / credentials.* |
| 9 | 入口文件 | *Application.java / main.py / app.js / main.go |
| 10 | 项目用途 | README 第一段或入口文件注释 |
检查点:
确认 10 项全部有值（非空、非"未知"）。如果有缺失，重新读取相关文件。
全部有值后，向用户展示探测结果确认。
阶段 2: Plan — 确定生成清单
使用下表（逐行问自己：这个项目需要吗？）：
项目有构建工具? ──── 是 → 生成 /build + /run + compile-check Hook
│
项目有测试目录? ──── 是 → 生成 /test 命令
│
项目有数据库/API/密钥? ──── 是 → 生成 security.md 规则 + block-sensitive Hook
│
项目有明确开发模式? ──── 是 → 生成 add-<feature> Skill
│ (如 @McpTool / Controller→Service→Repo / React 组件)
项目有框架特有审查需求? ──── 是 → 生成 specialist Agent
│
编译型语言 + 有构建工具 + 无测试? ──── 是 → 生成 compile-check Hook（强烈推荐）
| 文件 | 条件 | 优先级 |
|---|---|---|
CLAUDE.md |
始终 | 必须 |
.claudeignore |
始终 | 必须 |
.claude/settings.json |
始终 | 必须 |
.claude/rules/<lang>-conventions.md |
有代码时 | 建议 |
.claude/rules/security.md |
有数据库/API/密钥时 | 建议 |
.claude/commands/build.md |
有构建工具 | 建议 |
.claude/commands/run.md |
有启动命令 | 建议 |
.claude/skills/<add-feature>/SKILL.md |
有明确开发模式 | 可选 |
.claude/commands/test.md |
有测试目录 | 可选 |
.claude/agents/<specialist>.md |
有框架特有审查 | 可选 |
.claude/hooks/block-sensitive.ps1 |
有敏感操作 | 建议 |
.claude/hooks/compile-check.ps1 |
编译型+有构建 | 强烈建议 |
.gitignore 补充 |
缺标准排除 | 建议 |
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 419 lines · 103 tokens per session scan D 6de0d2a9a236
project-engineering-init is a skill published in the GitHub repository flasherses/claude-code-dotfiles (5 stars, last pushed 23d ago), licensed MIT. It adds 103 tokens to every session and 4,786 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it D with 3 findings (reads agent configuration directories, recursive force delete, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
paperclip
Interact with the Paperclip control plane API for task coordination and governance. Use when checking assignments, updating issue status, posting comments, delegating work, managing routines, or calling Paperclip API endpoints.
parallel-feature-development
Coordinate parallel feature development with file ownership strategies, conflict avoidance rules, and integration patterns for multi-agent implementation. Use this skill when decomposing a large feature into independent work streams, when two or more agents need to implement different layers of the same system…
mission-control
Interact with Mission Control — AI agent orchestration dashboard. Use when registering agents, managing tasks, syncing skills, or querying agent/task status via MC APIs.
teammode
Codex-only team orchestration: run a named team of cooperating Codex workers with durable, script-managed state. MUST USE when the user asks Codex to create, run, coordinate, inspect, archive, or delete a team of agents/threads/sessions, or to work on something as a team in parallel. FIRST inspects the active tool…
omh-agent-board
This is a Hermes-native agent-board workflow skill.
taiyi-ultrawork
TaiyiForge ultrawork — 并行切片 + Cursor Task 自动派发契约（对标 OMC ultrawork + spawnagent）.