Borrowing it
Nothing to install: this file belongs to fitlab-ai/agent-infra. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/fitlab-ai/agent-infra/main/AGENTS.mdgit clone --depth 1 https://github.com/fitlab-ai/agent-infraWrote 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/fitlab-ai/agent-infra/agents-md)<a href="https://agentmods.dev/instructions/fitlab-ai/agent-infra/agents-md"><img src="https://agentmods.dev/badge/instructions/fitlab-ai/agent-infra/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 5.1 | $0.03096 | $0.03096 |
| Opus 5 | $0.01548 | $0.01548 |
| Sonnet 5 | $0.00619 | $0.00619 |
| Haiku 4.5 | $0.00310 | $0.00310 |
Grade C, and why
agent-infra AGENTS.md scanned grade C with 1 finding 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 3d 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
- **不可逆 / 已发布的副作用**:已执行的破坏性操作(`rm -rf`、`git push --force`、`gh issue edit` 等)、跨轮次的设计决策与裁决理由。 How it starts
The opening of the file, as written. The whole thing — 207 lines — stays where its author put it; the contents beside it link to each section on GitHub.
agent-infra - AI 开发指南
本仓库包含 agent-infra 模板和技能仓库,用于多 AI 协作基础设施。
AI 行为准则(必须遵守)
源自 Andrej Karpathy 总结的 4 条 LLM 编程铁律,本项目所有 AI 工具均需遵守。 与下方项目规范冲突时,以项目规范为准;其余场景以本节为准。 这些准则偏向「稳」而非「快」,琐碎任务可酌情判断。
SKILL 执行场景的特例:在执行任一 SKILL 时,优先遵循 .agents/rules/no-mid-flow-questions.md(默认禁言 + 该规则文件列出的例外)。每次执行 SKILL 前应先 Read 该规则文件,以加载完整例外清单和具体约束。这与下文第 1 条「不确定就提问」不矛盾——SKILL 执行有明确的输入、输出和产物,不确定项应按最稳健方案推进并写入产物的「假设」/「未决问题」段落,由用户在审查检查点统一处理,而不是中途打断对话。
1. 先思考,再动手(Think Before Coding)
不要硬猜,不要藏起困惑,把权衡点摆到台面上。
- 显式声明你的假设;不确定就提问,不要默默猜测。
- 存在多种解释时,列出选项让用户选,不要擅自选定。
- 有更简单的方案就说出来,必要时反推用户的决定。
- 有任何不清楚的地方就停下来,指出困惑点并提问。
2. 简洁优先(Simplicity First)
只写解决当前问题所需的最少代码,不做任何投机性扩展。
- 不添加未被要求的功能、抽象、配置项。
- 不为单次使用的代码引入抽象层。
- 不为不可能发生的场景写错误处理。
- 写了 200 行但 50 行就够时,重写它。
- 自检:"资深工程师会觉得这过度设计吗?"——会,就简化。
兼容性默认关闭(Compatibility by Exception)
没有明确兼容承诺时,只实现当前契约,不主动保留旧行为。
- 不因“可能有旧调用方”而新增 adapter、wrapper、shim、双写、旧 schema 读取或迁移分支。
- 兼容性必须有证据:明确的旧消费者或存量数据、不可直接切换的原因、支持期限和删除条件;缺少任一项时按当前版本直接切换。
- 优先一次性边界迁移或可操作的失败提示,不在主路径长期维护新旧两套状态机。
- 临时兼容集中在唯一边界,并在该边界用
TODO(compat): Remove ... once ...标记删除对象和可验证删除条件;任务或 Issue 仅用于可选排期,不是兼容事实源或完成门禁。 - 当前流程的崩溃恢复、幂等重试和安全校验不等于历史兼容;只保留证明当前操作正确所需的最小事实。
- 发现任务范围外的既有兼容逻辑时只记录,不顺手删除;新增或延长兼容逻辑则必须遵循
.agents/rules/compatibility-policy.md。
3. 外科手术式修改(Surgical Changes)
只动该动的地方;只清理你自己制造的垃圾。
- 不顺手"优化"邻近代码、注释、格式。
- 不重构没坏的东西。
- 保持现有风格,即使你个人偏好不同。
- 发现无关的死代码——提醒用户,不要擅自删除。
- 因你的改动而产生的孤儿 import/变量/函数,必须由你清理。
- 自检:每一行改动都能直接追溯到用户的请求吗?
4. 目标驱动执行(Goal-Driven Execution)
先定义可验证的成功标准,再循环执行直到达成。
把任务转成可验证的目标:
- "加校验" → "先写非法输入的失败用例,再让它通过"
- "修 Bug" → "先写复现该 Bug 的用例,再修到通过"
- "重构 X" → "确保重构前后测试都通过"
多步任务先列计划,每一步都附带 verify 检查点:
1. [步骤] → verify: [检查项]
2. [步骤] → verify: [检查项]
3. [步骤] → verify: [检查项]
强成功标准让你能独立闭环;弱标准("让它跑起来")会让用户不断澄清。
这套准则生效的迹象:diff 中不必要的改动变少;因过度设计被推翻重写的次数变少;澄清问题发生在动手之前而不是踩坑之后。
上下文压缩规则(必须遵守)
当上下文窗口接近上限、需要对历史消息做有损压缩时,按以下优先级保留 / 丢弃信息。 本节约束 LLM 自身的压缩行为,不属于「项目规范」范畴,与上文 AI 行为准则并列、不冲突。
压缩时必须保留
- 任务上下文:当前
task-id、关联 Issue/PR 号、current_step、最近一次 SKILL 产物文件名(如analysis-r2.md、plan.md)。 - 用户偏好与显式 feedback:被告知「不要这样做」/「一直这样做」的指令、memory 风格的偏好声明。
- 不可逆 / 已发布的副作用:已执行的破坏性操作(
rm -rf、git push --force、gh issue edit等)、跨轮次的设计决策与裁决理由。 - 未结清的失败信号:未确认的报错栈、用户尚未回应的提问、被挂起的未决问题。
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.
- 3d ago First seen · 207 lines · 3,096 tokens per session scan C 3ae59d813dc0
agent-infra AGENTS.md is an instructions file published in the GitHub repository fitlab-ai/agent-infra (83 stars, last pushed yesterday), licensed MIT. It adds 3,096 tokens to every session, about $0.0155 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other instructions, from other repositories
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.
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).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
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.