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/lod-dawn/fast-vben-admin/agents-mdgit clone --depth 1 https://github.com/LoD-Dawn/Fast-Vben-AdminWrote 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/lod-dawn/fast-vben-admin/agents-md)<a href="https://agentmods.dev/instructions/lod-dawn/fast-vben-admin/agents-md"><img src="https://agentmods.dev/badge/instructions/lod-dawn/fast-vben-admin/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 | $0.02503 | $0.02503 |
| Opus 5 | $0.01252 | $0.01252 |
| Sonnet 5 | $0.00501 | $0.00501 |
| Haiku 4.5 | $0.00250 | $0.00250 |
Grade A, and why
Fast-Vben-Admin 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 6d 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 — 137 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Fast Vben Admin Agent Development Standard
本文件是所有自动化编码模型、代码生成模型和人工协作者的仓库级开发规范。进入任何子目录工作时,还必须读取并遵守距离目标文件最近的 AGENTS.md。子目录规范只能补充本文件,不能放宽本文件中的架构、安全和验证要求。
- 规范版本:1.0
- 架构基线:
docs/modular-architecture-implementation.md - 维护方式:架构变更与相关
AGENTS.md在同一变更中更新
1. 开始工作前
在修改代码前,按顺序完成:
- 阅读本文件和目标目录下的
AGENTS.md。 - 阅读与任务相关的事实源、ADR、现有实现和测试。
- 执行
git status --short,识别并保留用户已有修改。 - 用
rg定位现有实现、调用方、测试和兼容入口。 - 明确变更边界、数据所有者、迁移影响、Edition 影响和验证范围。
不要只依据需求描述、文件名或旧对话推断当前实现。代码、迁移、CI 和当前文档才是工作区事实。
2. 决策优先级
发生冲突时按以下顺序处理:
不要直接改写已经 Accepted 的 ADR 历史结论。需要改变已接受决策时,新增 ADR 并声明替代关系。Proposed ADR 不能由模型自行改成 Accepted;必须有团队签署和文档要求的演练证据。
3. 固定架构
- 架构形态是“模块化单体 + 构建期 Edition 组合”。当前阶段不拆微服务、不引入运行时微前端、不维护产品分支。
platform是每个 Edition 必选的交付模块。kernel、system、infra只是 Platform 内部限界上下文,不得加入 Edition YAML 或单独授权。items是业务模块样板。新增 IOA、ERP 等模块必须复用同一种模块契约、迁移、公开 API、事件和前端装配方式。- 新业务能力进入
backend/app/modules/<module_code>和frontend/apps/web-antd/src/modules/<module_code>,不得继续扩大中央业务模型、中央 Router 或全局业务页面。 - 业务模块只能依赖 Platform 稳定公开接口和其他模块的
public_api。禁止导入其他模块的domain、application、infrastructure、routes或 ORM 实体。 - 可选能力通过 capability 契约和组合根选择 Provider。禁止从可选模块实现建立静态依赖。
4. 唯一事实源
| 信息 | 人工维护的唯一事实源 | 禁止做法 |
|---|---|---|
| Edition 模块集合 | editions/<edition>.yaml |
用环境变量或菜单临时增加模块 |
| 模块契约 | ModuleDefinition |
在多个组合根重复声明 Router、权限或事件 |
| 构建组合 | 生成的 build-manifest.json |
前后端相互覆盖 Manifest |
| 模块期望/实际状态 | ModuleRegistry 与迁移/健康检查 |
请求鉴权时同步或创建运行时记录 |
| 套餐权益/例外/租户偏好 | TenantPlanModule / TenantModuleEntitlementOverride / TenantModule |
通过菜单或表存在性推断模块可用 |
| 用户权限 | Permission、Role、Menu 关系 | 只依赖前端按钮隐藏 |
| 数据表所有权 | Platform migration metadata 或模块自己的 migration namespace | 多模块共同拥有同一张表 |
派生文件必须通过仓库命令生成,不得手工维护生成结果。
5. 多租户与安全红线
- 租户身份只能来自已验证 JWT、可信事件信封、受控 client-to-tenant resolver、Schedule 的平台枚举结果或显式受审计的平台入口。
- 不得直接信任 body、query、header 中的
tenant_id。公开回调必须先通过租户编码、客户端凭据或其他可信映射解析租户,再进入 UoW。 - Tenant-owned 数据必须在对应 TenantUnitOfWork 内访问。API、Worker、Schedule、CLI 使用相同隔离原则。
privileged=True仅用于已经完成平台级授权的受控跨租户操作。调用点必须可审计,并具有跨租户负向测试。- ORM tenant 过滤不是 PostgreSQL RLS 的替代品;RLS 也不是可信租户解析的替代品。扩展 RLS 时必须同时更新模型集合、迁移、runtime 角色测试和调用路径。
- 不得假设所有带
tenant_id的历史表已经启用 RLS。OAuth2/OIDC、租户全局目录和全局模块运行时表的前置设计见实施基线和 ADR-0013。 - 运行时进程使用无超级用户、无建库/建角色、无
BYPASSRLS的app_runtime。迁移管理员凭据不得注入 API 或 Worker。 - 禁止提交密钥、真实密码、访问令牌、个人数据或生产连接信息。
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.
- 6d ago First seen · 137 lines · 2,503 tokens per session scan A f1dba4da2761
Fast-Vben-Admin AGENTS.md is an instructions file published in the GitHub repository LoD-Dawn/Fast-Vben-Admin (236 stars, last pushed 1mo ago), licensed MIT. It adds 2,503 tokens to every session, about $0.0125 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
fastapi-react-admin CLAUDE.md
Claude Code instructions for yilecoding/fastapi-react-admin, covering claude.md, 按任务导航, 让这套文档不腐烂:pnpm ctx:check, 这套文档怎么自己长大 and 结构.
fast-soy-admin AGENTS.md
Instructions for sleep1223/fast-soy-admin, covering fastsoyadmin - claude code guide, 重要文件(动手前先看), 标准操作流程, 任何修改之前 and 提交 / 推送之前.
fast-soy-admin CLAUDE.md
Instructions for sleep1223/fast-soy-admin, a project described as: 基于 FastAPI+Vue3+Naive UI 的现代化轻量管理平台。 A modern Management Platform based on FastAPI+Vue3+Naive UI.
tiptap AGENTS.md
AGENTS.md instructions for ueberdosis/tiptap, covering tiptap, rules, before opening a pr, code style and files.
adminforth AGENTS.md
AGENTS.md instructions for devforth/adminforth, covering agents.md, package manager, package names rules, general engineering rules and trust the contract.
geolens CLAUDE.md
Claude Code instructions for geolens-io/geolens: This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.