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/dengmengmian/muxlayer/agents-mdgit clone --depth 1 https://github.com/dengmengmian/muxlayerWrote 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/dengmengmian/muxlayer/agents-md)<a href="https://agentmods.dev/instructions/dengmengmian/muxlayer/agents-md"><img src="https://agentmods.dev/badge/instructions/dengmengmian/muxlayer/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 | $0.00817 | $0.00817 |
| Opus 5 | $0.00409 | $0.00409 |
| Sonnet 5 | $0.00163 | $0.00163 |
| Haiku 4.5 | $0.00082 | $0.00082 |
Grade A, and why
muxlayer 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.
What it actually says
核心原则
- 第一性原理：先判断真实目标，方向错就直接说，给更优路径。
- 极简沟通：中文输出、结论先行、语言直接，不复读背景。
- 简单优先：能小改不大改、能直观不抽象。
- 只改必要范围：每一处改动都能对应到本次需求。
- 暴露问题：发现风险、冲突、不可验证点，直说，不要糊。
- 不擅自开分支、不擅自重构、不擅自删除已有代码。
- 改后自检：是否改多、是否复杂化、是否漏验证、是否有更简方案。
工程规范
- 不捏造数据；生产代码禁止 Mock；Mock 只用于测试或本地调试。
- 不写用户没要求的功能、不做单次使用的抽象、不重构无关代码。
- 自己造成的 unused import / variable / 孤儿代码必须清理。
- 静默吞错、假成功、猜测式补丁、掩盖根因的兜底都禁止。
输出规范
- 结论先行。
- 对比、评审、任务拆解优先用 Markdown 表格。
- 简单问题直接答，不强行表格、不流水账、不用 P0/P1/P2。
TDD 模式
每个改动按以下顺序：
- 写测试：先写一个最小可失败的测试，明确预期行为。
- 跑红：确认测试因为目标行为缺失而失败，不是因为编译或环境错。
- 写实现：写最小代码让测试通过，不超出当前测试范围。
- 跑绿：所有测试通过，不只是新写的那一个。
- 重构：在测试保护下整理代码，每次小步、每次跑测试。
边界
- 不允许跳过红→绿步骤，不允许「先写实现回头补测试」。
- 不要为了凑覆盖率写无意义断言。
- UI / 端到端流程允许用手动验证替代单测，但要在 PR 里说明。
- 修 bug 必须先用一个能复现的测试锁住，再修。
代码评审模式
阅读 diff 时优先按以下顺序：
- 正确性：逻辑、边界、并发、错误路径是否正确。
- 安全：输入校验、注入、权限、密钥泄露。
- 性能：是否有 N+1、不必要的循环、阻塞调用。
- 可读性：命名、结构、注释是否帮助下一个读者。
- 风格：放最后，能格式化解决的不要单开评论。
反馈方式
- 区分 必改 / 建议 / 疑问，不要一律「建议」。
- 必改：指出问题 + 给出修复方向，不要只说「这样不好」。
- 建议：解释 trade-off，让作者自己决定。
- 疑问：表明自己不确定，不要假装是必改。
禁止
- 不要逐行复述代码做了什么。
- 不要在评审里夹带个人偏好（tabs vs spaces 之类）。
- 不要要求作者做超出 PR 范围的重构。
代码提交
- 所有的commit都要简洁，以用户视角来写。
- 没有说发版和推送就不要随意推到远程。
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 · 72 lines · 817 tokens per session scan A f10bf9d2957e
muxlayer AGENTS.md is an instructions file published in the GitHub repository dengmengmian/muxlayer (31 stars, last pushed 3d ago), licensed MIT. It adds 817 tokens to every session, about $0.0041 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
codex-router AGENTS.md
AGENTS.md instructions for duolahypercho/codex-router, covering model router installation instructions, repository maintenance workflow, choose the target, codex outcome and codex procedure.
codex-router CLAUDE.md
Claude Code instructions for duolahypercho/codex-router: Open and read AGENTS.md in full before installing, configuring, or extending this repository. It is the canonical instruction file for both Codex and Claude-based agents, including the procedures for adding local curated models, shipping repository-wide model…
deepseek-vision-skill CLAUDE.md
Claude Code instructions for reF0o0/deepseek-vision-skill, covering claude.md, 项目概述, 目录结构, 常用命令 and 架构要点.
cherry-studio AGENTS.md
AGENTS.md instructions for CherryHQ/cherry-studio, a project described as: AI productivity studio with smart chat, autonomous agents, and 300+ assistants. Unified access to frontier LLMs.
Tianshu-harness CLAUDE.md
Claude Code instructions for huiliyi37/Tianshu-harness, covering 天枢 (tianshu) / rivet, build & test, architecture, conventions and known constraints.
1flowbase AGENTS.md
AGENTS.md instructions for taichuy/1flowbase, covering 记忆, 用户偏好, 本项目相关skill在, 本项目 skills and 质量控制.