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/whynusn/typetype/agents-mdgit clone --depth 1 https://github.com/whynusn/typetypeWrote 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/whynusn/typetype/agents-md)<a href="https://agentmods.dev/instructions/whynusn/typetype/agents-md"><img src="https://agentmods.dev/badge/instructions/whynusn/typetype/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 | $0.15755 | $0.15755 |
| Opus 5 | $0.07878 | $0.07878 |
| Sonnet 5 | $0.03151 | $0.03151 |
| Haiku 4.5 | $0.01576 | $0.01576 |
Grade A, and why
typetype 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 4d 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.
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 — 575 lines — stays where its author put it; the contents beside it link to each section on GitHub.
typetype 项目开发指南
📍 文档导航卡（你在这里）
本文档面向 AI 开发者，记录编码规范和已知陷阱。
| 当前文档 | 其他核心文档 | 快速链接 |
|---|---|---|
| 本文 — 开发约束、编码规范、已知陷阱 | README.md — 快速入门ARCHITECTURE.md — 架构权威 | 代码风格已知陷阱 |
🧭 AI 阅读顺序
- 本文档 §3 代码风格 + §8 已知陷阱 — 编码约束和常见坑位
- ARCHITECTURE.md — 架构分层、数据流、依赖规则
- docs/reference/README.md — 配置/QML/API 速查
- 本文档 §4-7 — 测试策略、平台权限、CI
- docs/history/ — 历史功能设计文档（已完成，仅作背景参考）
已熟悉项目后，日常开发只需查阅 §3 代码风格 + §8 已知陷阱 即可。
📚 文档维护指南
文档职责速查
| 文档 | 角色 | 什么时候更新 |
|---|---|---|
ARCHITECTURE.md |
架构事实来源 | 新增/删除文件、架构变更、新架构陷阱 |
| 本文档 | AI 开发约束与陷阱集 | 新坑位、编码规范变化、验证要求更新 |
docs/reference/* |
速查表 | 配置字段、Bridge Slot、API 端点变化 |
docs/decisions/* |
架构决策记录（ADR） | 做出重大架构决策 |
docs/history/* |
历史设计文档归档（冻结） | 完成重大功能、修复复杂 bug |
CHANGELOG.md |
发布历史 | 版本发布或用户可见变更 |
权威矩阵
摘要；完整定义（含跨维度冲突处理）见 docs/meta/README.md。
事实可靠性链：源码 > ARCHITECTURE.md > reference/* > decisions/* > AGENTS.md > guides/* > history/*
操作优先级链：AGENTS.md > guides/* > ARCHITECTURE.md > decisions/* > reference/* > history/*
修改代码后的文档更新流程
改完代码
↓
判断变更类型（见上表；逐项映射见 docs/meta/README.md § 同步规则）
↓
更新对应文档
↓
验证：运行 docs/meta/README.md § 验证清单
提交前验证清单
-
ARCHITECTURE.md目录结构与src/backend/实际文件一致 -
docs/reference/中的表格与代码一致 -
ARCHITECTURE.md陷阱覆盖最新发现 - 所有内部链接无断链
- 本文档陷阱描述准确
-
CHANGELOG.md已更新（若涉及用户可见变更）
1. 开发环境与命令
字体裁剪
| 字体 | 原始大小 | 裁剪后大小 | 减少 |
|---|---|---|---|
| HarmonyOS Sans SC Regular | 8.2 MB | 504 KB | ~94% |
| LXGW WenKai Regular | 25.4 MB | 880 KB | ~97% |
打包时使用 *-subset.ttf 而非原始字体。
💬 用户快捷操作指令
| 用户说 | AI 执行 |
|---|---|
| "项目概览" | 阅读 README 摘要 + ARCHITECTURE 一句话理解，汇总输出 |
| "同步文档" | 按"代码变更→文档更新映射"逐项检查更新 |
| "检查文档" | 运行 scripts/verify-framework.sh，汇总结果 |
| "记录决策" | 在 docs/decisions/ 创建 ADR |
| "更新 CHANGELOG" | 检查 git 提交，追加版本条目 |
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.
- 4d ago First seen · 575 lines · 15,755 tokens per session scan A 155321d12a52
typetype AGENTS.md is an instructions file published in the GitHub repository whynusn/typetype (11 stars, last pushed 16d ago), licensed MIT. It adds 15,755 tokens to every session, about $0.0788 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 2026-08-30.
Other instructions, from other repositories
nourish CLAUDE.md
Instructions for y5-snowies/nourish, covering claude.md, what this is, crate dir that produces the y5compositor binary, build, check & run and multi-workspace architecture (read this first).
unisic CLAUDE.md
Claude Code instructions for unisic/unisic, covering claude.md, build & run, non-negotiables, what unisic is and planning workflow.
unisic AGENTS.md
AGENTS.md instructions for unisic/unisic, covering agents.md, 1. what unisic is (and is not), non-negotiable product constraints, 2. prime directives and 3. build, run, and dependencies.
Pake copilot-instructions.md
Copilot instructions for tw93/Pake, a project described as: 🤱🏻 Turn any webpage into a desktop app with one command.
chat-on-steroids CLAUDE.md
Claude Code instructions for totec448-spec/chat-on-steroids: Read and follow AGENTS.md before changing this repository.
agent-zero AGENTS.md
AGENTS.md instructions for agent0ai/agent-zero, covering agent zero dox, purpose, project, root ownership and project-wide contracts.