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/yangfan-code-star/context-dev/agents-mdgit clone --depth 1 https://github.com/YangFan-Code-Star/context-devWhat 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 | $0.02651 | $0.02651 |
| Opus 5 | $0.01326 | $0.01326 |
| Sonnet 5 | $0.00530 | $0.00530 |
| Haiku 4.5 | $0.00265 | $0.00265 |
Grade A, and why
context-dev 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 yesterday.
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 — 110 lines — stays where its author put it; the contents beside it link to each section on GitHub.
context-dev — Agent 工作手册
这是框架本身的仓库，不是用它初始化出来的项目。 .agents/skills/ 是完整版产品——五个要被安装到别人项目里的技能；lite/ 是轻量版发行包——两个技能 + 四个骨架文件，供个人 / 中小型项目长期迭代使用。两者里的 templates/ 都是产出物，是别人项目的起点。把它们当成本仓库的配置来改，是这里最容易犯、也最贵的错。
本文件不会被两份 scaffold.mjs 中的任何一份复制到别处（它们只复制各自的 templates/）；但 GitHub Template 会把它复制进新仓库，此时 scaffold.mjs 会识别上面的 FRAMEWORK-AGENTS 标记并在 README 仍带 FRAMEWORK-README 标记时受控替换。改它只影响在本仓库里干活的 agent。
快速上手
零依赖，无需安装，只要 Node 20+（CI 固定 20）：
node --version
node --test
验证命令（改完代码必须自己跑，不要让用户当第一个发现问题的人）：
| 目的 | 命令 |
|---|---|
| 测试（脚手架与体检脚本的冒烟） | node --test |
| 语法检查 | node --check .agents/skills/init/scaffold.mjs、node --check .agents/skills/init/templates/scripts/audit-context.mjs、node --check lite/scaffold.mjs |
| 体检模板树（看输出用） | node .agents/skills/init/templates/scripts/audit-context.mjs |
| 体检仓库根（CI 同款） | node .agents/skills/init/templates/scripts/audit-context.mjs --root . |
语法检查里的三个文件、体检仓库根和 node --test 合起来就是 .github/workflows/ci.yml 的全部内容，没有额外的全量档。
体检脚本本身已被 node --test 覆盖，手动跑只是为了看输出。不带 --root 时 ROOT 由脚本自身位置推导，所以检查的是 templates/ 子树而不是仓库根；--root . 才会体检本仓库根。模板树的合格状态是「0 error、若干占位符 warning」——占位符是 warning 不是 error。仓库根体检的「缺少 docs/learning-inbox.md」warning 是预期：根目录按铁律 4 不补 docs/，学习闭环只存在于 templates/ 生成的用户项目里。因此 /ship-change 第 7 步在本仓库没有收件箱可写：蒸馏信号直接并入本文件或 README，不要创建 docs/。
目录地图
AGENTS.md 本文件。每次对话全量加载
.agents/skills/ 完整版产品：五个技能，完整版唯一需要安装到用户项目的东西
init/SKILL.md 分阶段访谈流程（含轻量档）
init/scaffold.mjs 把 templates/ 生成到用户项目根
init/references/ 访谈题库与答案落盘映射，按需加载
init/templates/ 完整版骨架模板 —— 产出物，不是本仓库的配置
lite/ 轻量版发行包：init + maintain 两个技能、四个骨架文件
README.md 轻量版安装与用法，完整版 / lite 二选一
scaffold.mjs 把 lite/templates/ 生成到用户项目根
skills/ 两个轻量技能，安装到用户项目 .agents/skills/
templates/ 轻量骨架模板 —— 产出物，不是本仓库的配置
tests/ scaffold 与 audit 的冒烟测试，脚本的行为边界以它为准
.github/workflows/ci.yml 语法检查 + node --test
README.md 对外说明书，带 FRAMEWORK-README 标记（初始化时被整份替换）
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.
- yesterday First seen · 110 lines · 2,651 tokens per session scan A 308d46b21751
context-dev AGENTS.md is an instructions file published in the GitHub repository YangFan-Code-Star/context-dev (2 stars, last pushed 14d ago), licensed MIT. It adds 2,651 tokens to every session, about $0.0133 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
booster_mcp AGENTS.md
AGENTS.md instructions for NeuroGhostDev/booster_mcp, covering booster agent bootstrap, bootstrap contract, engineering workflow, context priority and important boundary.
Styler AGENTS.md
AGENTS.md instructions for TidyFactor/Styler, covering agents.md — tidyfactor styler workspace rules and key rules.
finding-unknowns-skills CLAUDE.md
Claude Code instructions for Neeeophytee/finding-unknowns-skills, covering repo notes, what ships, and the gate that decides, version lives in three manifests — move them together, files that must be edited together and shipped guidance is not this file.
vvvv-skills CLAUDE.md
Claude Code instructions for tebjan/vvvv-skills, covering vvvv gamma, when to patch vs code, core concepts, live environment and important conventions.
pi AGENTS.md
AGENTS.md instructions for counterposition/pi, covering agents.md, commands, pattern: pnpm --filter @counterposition/pi-web-search run, single test file and single test by name.
hyperskills CLAUDE.md
Instructions for hyperb1iss/hyperskills, a project described as: My collection of agent skills for a variety of tasks.