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 instructions/int2t05/engineering-skills/agents-mdgit clone --depth 1 https://github.com/int2t05/engineering-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/int2t05/engineering-skills/agents-md)<a href="https://agentmods.dev/instructions/int2t05/engineering-skills/agents-md"><img src="https://agentmods.dev/badge/instructions/int2t05/engineering-skills/agents-md.svg" alt="Measured on agentmods" height="20"></a>What 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 | $0.02421 | $0.02421 |
| Opus 5 | $0.01210 | $0.01210 |
| Sonnet 5 | $0.00484 | $0.00484 |
| Haiku 4.5 | $0.00242 | $0.00242 |
Grade A, and why
engineering-skills AGENTS.md scanned grade A with 0 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 today.
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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 132 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Universal entry point for AI coding agents working with this skill pack — Claude Code,
Codex, OpenCode, Cline, Continue, Cursor (agents mode), and any agent that reads
AGENTS.md.
Scope: This file orients agents using the
engineering-skillspack. The reusable assets are the 47 skills underskills/; this file routes work to them and loads the shared discipline.
What this pack is
47 skills organized by the software development lifecycle, including a meta router.
Each skill is a folder skills/<phase>/<name>/SKILL.md with YAML frontmatter (name,
description, optional disable-model-invocation) and four sections: When to use / Steps /
Verify / References. Encyclopedic data lives in per-skill references/ (progressive
disclosure).
Read the engineering principles first
Before any task, load references/engineering-principles.md
(repo-relative path — resolves in every framework). It holds the 9 shared principles every
skill assumes: surface assumptions, manage confusion, push back, enforce simplicity, surgical
scope, verify don't assume, plan with built-in plan mode, goal-driven execution, behavior-preserving
change discipline.
Claude Code gets these injected automatically at session start (SessionStart hook). Other frameworks — read the file above once per session.
Intent → skill routing
When a task arrives, identify the phase and reach for the matching skill. If unsure, the
using-skills skill (in skills/meta/) maps the task to a phase.
| Phase | Skill | Triggers on |
|---|---|---|
| product | brainstorm |
underspecified ask, "refine this idea", "头脑风暴" (user-typed) |
| product | spec |
new project/feature/change, "write spec", "create prd", "写需求文档", "要做什么", "需求是什么" |
| product | oss-strategy |
OSS business model, COSS, "开源策略", "开源商业模式" |
| research | research |
deep web research, cited report, "深度检索", "调研报告" |
| research | research |
"deep research", "市场调研", "竞品分析", "技术选型", "选哪个" (general/market/tech-selection modes) |
| design | architecture |
system design, "架构设计", "系统设计", ADR |
| design | domain-modeling |
domain model, ubiquitous language, "领域模型", CONTEXT.md |
| design | api-design |
REST/GraphQL contracts, "接口设计", "API 契约" |
| design | codebase-design |
deep modules, "深化模块", "重构架构" |
| design | frontend-design |
UI, "界面设计", "前端设计", production-grade frontend |
| design | image-to-code |
"设计图转代码", "图片实现", image-first frontend pipeline |
| design | imagegen |
"品牌识别", "logo 设计", "网站设计图", "移动端设计图", brand/web/mobile design reference images |
| design | design-research |
"设计参考", "真实设计参考", "找设计灵感", free public design galleries |
| design | schema-design |
"数据模型", "表结构设计", data model, schema design |
| design | prompt-engineering |
"提示词工程", "LLM 特性", prompt design, eval harness |
| design | prototype |
throwaway prototype, "原型", "试这个方案", "build a demo" |
| develop | implement |
implement spec/tickets, "实现", "编码", "改这个配置", "搭项目骨架", "改一下这个", "加个功能" |
| develop | multi-agent-orchestration |
"并行 agent", "子代理执行", "派发任务", parallel/sequenced subagents |
| meta | skill-authoring |
"写技能", "改技能", "技能评估", write/edit/eval a skill |
| develop | breakdown |
break work into tickets, "拆解任务", "拆票", decision map |
| develop | context-engineering |
agent needs context, "解释这段代码", "带我过一遍代码库" |
| develop | i18n |
"国际化", "本地化", "多语言", i18n, RTL, localization |
| develop | auth-implementation |
"implement auth", "实现认证", "实现登录", "OAuth 集成", "password hashing", "token 轮换", JWT, RBAC |
| develop | error-handling |
"error handling", "错误处理", "重试策略", "熔断", "降级", retry strategy, circuit breaker, error propagation |
| tune | cost-optimization |
"成本优化", "云账单", "费用", cloud bill, FinOps, spend |
| tune | performance |
"性能优化", "性能调优", "太慢了", "卡顿", profile, bottlenecks |
| tune | simplify |
"too complex", "简化", "重构求清晰", refactor for clarity |
| tune | refactoring |
"refactor structure", "提取模块", "拆分文件", extract/move/split/merge |
| test | tdd |
"测试驱动开发", "红绿重构", red-green-refactor |
| test | test-generation |
"生成测试", "补测试", generate test files |
| test | api-testing |
contract testing, "API 测试", "接口测试", REST/GraphQL |
| test | e2e-testing |
e2e, browser test, "端到端测试", "浏览器测试" |
| test | load-testing |
"压测", "压力测试", "容量测试", load/stress test, capacity |
| verify | a11y-review |
"无障碍", "可访问性", accessibility, a11y, WCAG, screen reader |
| verify | code-review |
review before merge, "代码审查", "合并前审查", "帮我看看这代码", "能合并吗" |
| verify | debugging |
bug, "调试", "排查 bug", "读日志", "排查错误日志", "为什么返回 null", "跑不通" |
| verify | security-review |
security review, "安全审查", "安全审计", secrets, injection |
| verify | linting |
"lint", "lint 修复", "静态分析", eslint/ruff/clippy |
| ship | shipping |
deploy, launch, "上线", "发布", "部署到生产" |
| ship | git-workflow |
commit, merge conflict, "提交", "合并冲突", pre-commit |
| ship | ci-cd |
CI pipeline, GitHub Actions, "流水线", "持续集成" |
| ship | deprecation-migration |
deprecate, migrate, "迁移", "升级依赖", "更新这个库" |
| ship | oss-polish |
"开源项目美化", "优化项目展示", README/topics polish |
| operate | observability |
logs, metrics, "可观测性", "监控告警", instrumentation |
| operate | documentation-audit |
docs drift, "写文档", "写 README", "文档化这个功能", sync docs |
| operate | purity-audit |
"纯净审计", "残留审计", files read like edited many times, historical residue |
| operate | handoff |
hand off to another session, "交接" (user-typed) |
| operate | incident-response |
"事故响应", "线上故障", "复盘", incident, on-call, postmortem |
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.
- today Changed · +2 lines · +54 tokens per session aba6457006c6
- 4d ago First seen · 130 lines · 2,367 tokens per session scan A 70e88e2156ae
engineering-skills AGENTS.md is an instructions file published in the GitHub repository int2t05/engineering-skills (3 stars, last pushed 3d ago), licensed MIT. It adds 2,421 tokens to every session, about $0.0121 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
tdder AGENTS.md
AGENTS.md instructions for t1/tdder, covering agents.md, file locations, claude code vs pi, development and dual-platform agents.
sdlc-graph-engineering CLAUDE.md
Instructions for RonMizrahi/sdlc-graph-engineering, covering sdlc-graph-engineering — repository guide, sources of truth, changing the skill — required, authoring rules and the skill's own standard, applied to itself.
dev-crew AGENTS.md
AGENTS.md instructions for morodomi/dev-crew, covering dev-crew, start here, tech stack, skills and quick start.
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.