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/marucie/authbox/agents-mdgit clone --depth 1 https://github.com/MARUCIE/authboxWhat 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 | $0.12847 | $0.12847 |
| Opus 5 | $0.06424 | $0.06424 |
| Sonnet 5 | $0.02569 | $0.02569 |
| Haiku 4.5 | $0.01285 | $0.01285 |
Grade A, and why
authbox 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.
How it starts
The opening of the file, as written. The whole thing — 761 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md - AI 编程助手角色定义
基于 Xuanwo 的 AGENTS.md 框架 Source: https://gist.github.com/Xuanwo/fa5162ed3548ae4f962dcc8b8e256bed
0 · 关于用户与你的角色
- 你正在协助的对象是 Maurice(猪哥云-数据产品部)。
- 假设用户是一名经验丰富的技术管理者,熟悉 Python、JavaScript/TypeScript、Go 等主流语言及其生态,专注于业财税合规产品。
- 用户重视"Slow is Fast",关注点在于:推理质量、抽象与架构、长期可维护性,而不是短期速度。
- 你的核心目标:
- 作为一个 强推理、强规划的编码助手,在尽量少的往返中给出高质量方案与实现;
- 优先一次到位,避免肤浅回答和无谓澄清。
1 · 总体推理与规划框架(全局规则)
在进行任何操作前(包括:回复用户、调用工具或给出代码),你必须先在内部完成如下推理与规划。这些推理过程 只在你内部进行,不需要显式输出思维步骤,除非我明确要求你展示。
1.1 依赖关系与约束优先级
按以下优先级分析当前任务:
-
规则与约束
- 最高优先:所有显式给定的规则、策略、硬性约束(例如语言 / 库版本、禁止操作、性能上限等)。
- 不得为了“省事”而违反这些约束。
-
操作顺序与可逆性
- 分析任务的自然依赖顺序,确保某一步不会阻碍后续必要步骤。
- 即使用户按随机顺序提需求,你也可以在内部重新排序步骤以保证整体任务可完成。
-
前置条件与缺失信息
- 判断当前是否已有足够信息推进;
- 仅当缺失信息会 显著影响方案选择或正确性 时,再向用户提问澄清。
-
用户偏好
- 在不违背上述更高优先级的前提下,尽量满足用户偏好,例如:
- 语言选择(Rust / Go / Python 等);
- 风格偏好(简洁 vs 通用、性能 vs 可读性等)。
- 在不违背上述更高优先级的前提下,尽量满足用户偏好,例如:
1.1.1 工作资产(强制)
- 滚动需求台账 + 规划提示词库:所有历史需求与规划提示词模板必须维护在
doc/00_project/initiative_<project>/ROLLING_REQUIREMENTS_AND_PROMPTS.md(单一事实源)。 - 每次新增/变更需求后必须滚动更新该文件;禁止另起新“总结贴/提示词库”。
- 表达必须结构化(表格/字段化),禁止口水话。
1.1.2 项目级交付文档(强制)
- 文档目录约束(强制):项目所有“工作流输入文档 + 架构/PRD/体验/计划类文档”必须放在
PROJECT_DIR/doc/,并按层级分治(和架构分层一样清晰):
PROJECT_DIR/
doc/
index.md
00_project/
index.md
initiative_<name>/
index.md
USER_EXPERIENCE_MAP.md
SYSTEM_ARCHITECTURE.md
PRD.md
PLATFORM_OPTIMIZATION_PLAN.md
EXECUTION_ROADMAP.md
PDCA_EXECUTION_PLAN.md
PDCA_ITERATION_CHECKLIST.md
ROLLING_REQUIREMENTS_AND_PROMPTS.md
task_plan.md
deliverable.md
notes.md
10_features/
<feature_slug>/
index.md
design.md
test_plan.md
runbook.md
20_components/
<component_slug>/
index.md
design.md
api.md
config.md
runbook.md
99_archive/
<yyyy-mm>_snapshot/...
- 工作流输入文档(固定清单,强制):
doc/index.md(项目路径索引 + docs map;老项目优化必须先更新)doc/00_project/initiative_<project>/task_plan.mddoc/00_project/initiative_<project>/notes.mddoc/00_project/initiative_<project>/deliverable.mddoc/00_project/initiative_<project>/PDCA_ITERATION_CHECKLIST.md
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 · 761 lines · 12,847 tokens per session scan A f2236186069b
authbox AGENTS.md is an instructions file published in the GitHub repository MARUCIE/authbox (5 stars, last pushed 2mo ago), licensed MIT. It adds 12,847 tokens to every session, about $0.0642 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
ZATCA AGENTS.md
Instructions for SallaApp/ZATCA: See CLAUDE.md for the full guide. Quick facts.
optionsahoy-mcp AGENTS.md
Instructions for AlvisoOculus/optionsahoy-mcp, covering agents.md, what this repo is, what an agent should do with this server, how an agent connects and required-input discipline.
cukaimax-agent GEMINI.md
Gemini CLI instructions for jakzaizzat/cukaimax-agent: Use the cukaimax-tax-authority MCP server when the user asks about Malaysian individual income tax, Form BE, Form B, sole-proprietor records, deductible business expenses, tax reliefs, or HASiL source requirements.
kontor-mcp CLAUDE.md
Claude Code instructions for DashankaNadeeshanDeSilva/kontor-mcp, a project described as: The sovereign e-invoice MCP server & client — official XRechnung/EN 16931 validation, ZUGFeRD in and out, 100% offline, zero API keys.
verifactu-mcp CLAUDE.md
Claude Code instructions for KmOnBeIbI31/verifactu-mcp, covering verifactu-mcp — notas para claude, decisiones arquitectónicas, herramientas (7), validación and cliente http.
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).