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/aser-ho/coding-agent-safety-gate/agents-mdgit clone --depth 1 https://github.com/ASER-ho/coding-agent-safety-gateWhat 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 | $0.00589 | $0.00589 |
| Opus 5 | $0.00295 | $0.00295 |
| Sonnet 5 | $0.00118 | $0.00118 |
| Haiku 4.5 | $0.00059 | $0.00059 |
Grade A, and why
coding-agent-safety-gate 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 2d 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
AGENTS / 代理规则
本文件由 AI 编程助手（Claude Code、Codex、Cursor、Gemini CLI 等）进入仓库时读取，定义它们必须遵守的安全约束。
This file is read by AI coding agents (Claude Code, Codex, Cursor, Gemini CLI, and others) when they enter this repository. It defines the safety constraints they must follow.
仓库类型 / Repository type
本仓库是文档优先的。只包含 Markdown 文件：一个 skill 定义、模板、文档和示例。没有运行时代码、没有 package.json、没有依赖、没有自动化。
This repository is documentation-first. It contains only Markdown files: a skill definition, templates, docs, and examples. There is no runtime code, no package.json, no dependencies, and no automation.
AI 代理规则 / Rules for AI coding agents
在本仓库工作时：
- 规则 #1 — 在违反本文件任何规则之前，必须先停下来，获得用户的明确批准。
- 优先做小的、仅 Markdown 的改动。
- 不要新增 CLI、MCP server、VS Code 插件、npm 包、runtime enforcement 工具、secret scanner、GitHub Actions 工作流、release 自动化或任何依赖。
- 不要读取、打印、推断、请求或存储真实机密（API key、token、cookie、私钥、凭据或 SecretStore 值）。
- 不要在任何文件中暴露完整的本地机器路径。
- 未经用户明确批准，不要执行
git push、git tag、创建 release、添加 remote、force push 或重写历史。 - 未经用户确认，不要决定下一步任务。
- 在声称完成工作之前提供证据。
- 如果任务范围不清晰，停止并标记
NEEDS HUMAN REVIEW。
When working in this repository:
- Rule #1 — stop and get explicit human approval before breaking any rule in this file.
- Prefer small, Markdown-only changes.
- Do not add a CLI, MCP server, VS Code plugin, npm package, runtime enforcement tool, secret scanner, GitHub Actions workflow, release automation, or any dependency.
- Do not read, print, infer, request, or store real secrets (API keys, tokens, cookies, private keys, credentials, or SecretStore values).
- Do not expose full local machine paths in any file.
- Do not run
git push,git tag, create a release, add a remote, force push, or rewrite history without explicit human approval. - Do not decide the next task without human confirmation.
- Provide evidence before claiming work is complete.
- If task scope is unclear, stop and mark
NEEDS HUMAN REVIEW.
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.
- 2d ago First seen · 38 lines · 589 tokens per session scan A 9e6d4336971b
coding-agent-safety-gate AGENTS.md is an instructions file published in the GitHub repository ASER-ho/coding-agent-safety-gate (2 stars, last pushed 1mo ago), licensed MIT. It adds 589 tokens to every session, about $0.0029 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
meta-llm-charter CLAUDE.md
Instructions for entropyvortex/meta-llm-charter, covering meta v3.1 core charter, bias, meta-0, r1 decompose and r2 ask gate.
awesome-reviewers CLAUDE.md
Instructions for baz-scm/awesome-reviewers, covering awesome reviewers — repository guidelines, what this project is, source of truth, site layer and machine interface.
code-context AGENTS.md
Instructions for infino-ai/code-context, covering code-context: notes for ai agents, project overview, repo map, build, test, gates and conventions.
gtm-cheat-codes CLAUDE.md
Instructions for zapier/gtm-cheat-codes, covering repo conventions and commits.
Stata-CLI AGENTS.md
Instructions for Utolaris/Stata-CLI, covering agents.md, project structure and working rules.
waybill AGENTS.md
Instructions for wardmos/waybill, covering agents.md, privacy, checks, commit messages and git history.