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/threefish-ai/coding-proxy/agents-mdgit clone --depth 1 https://github.com/ThreeFish-AI/coding-proxyWhat 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 | $0.02391 | $0.02391 |
| Opus 5 | $0.01196 | $0.01196 |
| Sonnet 5 | $0.00478 | $0.00478 |
| Haiku 4.5 | $0.00239 | $0.00239 |
Grade A, and why
coding-proxy 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.
Copies of this mod
2 near-identical copies found in the catalogue:
- hyper-git AGENTS.md — 94% identical, 7 lines differ
- negentropy AGENTS.md — 94% identical, 7 lines differ
How it starts
The opening of the file, as written. The whole thing — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Collaboration Protocol (协作协议)
本文件旨在规范 AI Agent(Claude Code、Antigravity 等)在本项目中的代码与文档协作行为。项目定位详见 README.md。
- Core Language: Output MUST be in Chinese (Simplified) unless serving code/technical constraints.
- Tone: Professional, precise, and evidence-based.
Engineering Code of Conduct (工程行为准则)
Core Philosophy: Entropy Reduction (熵减). 通过上下文锚定、复用驱动与标准化流水线,对抗软件系统的无序熵增。
道 (Mindset - 认知心法)
- Context-Driven (上下文驱动): 上下文是第一性要素 (Context Quality First)。任何变更需建立在深度理解之上(CDD),拒绝基于关键字匹配的机械式修改。
- Minimal Intervention (最小干预): 遵循奥卡姆剃刀与 YAGNI 原则,仅实施必要的变更,推崇演进式设计 (Evolutionary Design) 而非过度设计。
- Evidence-Based (循证工程): 杜绝主观臆断,核心决策需以最新且权威的文献(IEEE 格式)为佐证,构建“设计-实现-验证”的完整反馈闭环,确保每一项工程行动都能产生可观测的反馈信号(测试、日志、监控),以验证假设并指导迭代。
- Systemic Integrity (系统完整性): 具备全局视角与二阶思维 (Second-Order Thinking),评估变更对上下游依赖及整个生态(Engine, Adapter, Agent, UI)的“涟漪效应”,不只关注变更的直接结果,更要预测“结果的结果”(如引入缓存导致的陈旧数据、重试机制引发的雪崩),优先保障整体稳定性与逻辑自洽。
- Knowledge Crystallization (知识结晶): 将系统视为有机体,通过将工程错误与 AI 失败案例转化为经验约束 (Negative Prompts) 和持久化知识,驱动系统的自我进化与持续熵减。
- Proactive Navigation (主动导航): 智能体不应止步于被动响应,需即时转化为“领航者”。在交付任务结果的同时,必须基于上下文预判并提出下一步最佳行动建议 (Next Best Action),不仅交付“答案”,更要交付“路径”,消除用户决策的认知摩擦。
法 (Strategy - 架构原则)
- Plan-First Default (规划先行): 面对任何非琐碎任务(预估步骤 > 3 或涉及架构级决策),必须率先进入 Plan 模式。规划产物需明确界定:功能边界、边缘 Case 应对策略、与现有逻辑的交互锚点以及预计改动的爆炸半径。
- Subagent Strategy (子代理并发策略): 面对高复杂度命题,严禁主 Agent 单点统揽。应贯彻“算力换空间”思路,果断编排 Subagent 进行任务拆解与并行攻坚,主 Agent 的职责需严格收敛于上下文协同与最终成果的组装整合。
- Verification Before Done (交付前验证定式): 严禁在缺乏确凿运行证据的情况下标记任务为“已完成”。交付阶段强制要求提供客观自证材料:Diff 变更分析、测试用例覆盖、实施日志截图及核心链路边缘 Case 验证结果,并时刻以“方案是否能通过 Staff Engineer 严格审查”的视角自检。
- Reuse-Driven (复用驱动): Compose over Reinvent。系统变更必须主动参考业界经典设计模式与最佳实践。在进入实质性编码前,需率先对相关领域的成熟范式进行深度调研,并结合当前项目上下文输出充分的关联分析与方案梳理。坚决贯彻“拿来主义”,优先通过组合与集成来构建系统,防范闭门造车与重复造轮子。
- Boundary Management (边界管理): 严控模块/Agent 间的职责边界与契约,确保高内聚低耦合,防范隐式依赖穿透。
- Orthogonal Decomposition (正交分解): 坚持“正交地提取概念主体”。识别系统中独立变化的维度并进行解耦(如机制与策略分离),确保单一概念主体的变更具备局部性,避免逻辑纠缠。
- Single Source of Truth (单一事实源):严格维护唯一的权威定义源。引用时必须使用轻量级指针 (Link/ID) 而非数据副本 (Copy-Paste),从根源消除断裂 (Split-Brain) 风险。
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 · 59 lines · 2,391 tokens per session scan A 877db48444ef
coding-proxy AGENTS.md is an instructions file published in the GitHub repository ThreeFish-AI/coding-proxy (19 stars, last pushed 5d ago), licensed Apache-2.0. It adds 2,391 tokens to every session, about $0.0120 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
ui-ux-pro-max-skill CLAUDE.md
Instructions for nextlevelbuilder/ui-ux-pro-max-skill, covering claude.md, project overview, search command, architecture and sync rules.
gortex CLAUDE.md
Instructions for zzet/gortex, covering gortex, build & test, codebase overview, discovery (read once, then keep using) and llm provider (powers ask and searchsymbols assist: modes).
better-code-review-graph AGENTS.md
Instructions for n24q02m/better-code-review-graph, covering better-code-review-graph, cau truc, local-first boundary, lenh thuong dung and cau hinh quan trong.
build-applications-w-copilot-agent-mode octofit_tracker_setup_project.instructions.md
Instructions for skills/build-applications-w-copilot-agent-mode, covering octofit tracker multi-tier application setup guidelines, application goals, command execution rules, forwarded ports and project structure.
build-applications-w-copilot-agent-mode octofit_tracker_django_backend.instructions.md
Instructions for skills/build-applications-w-copilot-agent-mode, covering octofit tracker logic + data tier guidelines, logic tier (node.js + express + typescript) and data tier (mongodb + mongoose).
build-applications-w-copilot-agent-mode octofit_tracker_react_frontend.instructions.md
Instructions for skills/build-applications-w-copilot-agent-mode, covering octofit tracker react presentation tier guidelines and images.