Borrowing it
Nothing to install: this file belongs to jianchen08/Agent-os-open. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/jianchen08/Agent-os-open/main/AGENTS.mdgit clone --depth 1 https://github.com/jianchen08/Agent-os-openWrote 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/jianchen08/agent-os-open/agents-md)<a href="https://agentmods.dev/instructions/jianchen08/agent-os-open/agents-md"><img src="https://agentmods.dev/badge/instructions/jianchen08/agent-os-open/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 | $0.03156 | $0.03156 |
| Opus 5 | $0.01578 | $0.01578 |
| Sonnet 5 | $0.00631 | $0.00631 |
| Haiku 4.5 | $0.00316 | $0.00316 |
Grade A, and why
Agent-os-open 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.
How it starts
The opening of the file, as written. The whole thing — 141 lines — stays where its author put it; the contents beside it link to each section on GitHub.
灵汐 AgentOS — 项目记忆（供 AI 协作者）
本文件是给在本仓库工作的 AI 代理（DSH 等）的项目上下文。读完它再动手； 口径冲突时以本文件 +
docs/decisions/为准（README / ARCHITECTURE / guides 已对齐 0.2 现状）。
项目是什么
可进化的智能体操作系统：把 LLM、工具、记忆、任务、配置组织成可观测、可干预、 可回滚的管道（Pipeline）。当前主线为 0.2 架构：Rust 微内核 + Python 插件 + React 前端。
仓库布局（0.2 现实）
| 路径 | 内容 |
|---|---|
kernel/ |
Rust 微内核（crates：api/config/core/db-admin/engine/hooks/http/invoker/mcp/plugin-loader/session/tenant/user-admin；agentos-kernel 是主进程） |
plugins/ |
Python 插件（shared/system/ 系统插件、shared/tools/ 工具插件、shared/pipeline/ 管道插件） |
frontend/src/ |
React 19 + Vite + Zustand + Antd（RJSF v6 表单引擎） |
config/ |
运行配置（agents/tools/plugins/pipelines/isolation/storage.yaml 等） |
docs/ |
文档；docs/decisions/ = ADR 决策记录；docs/working/ = 工作文档与研究报告 |
tests/ |
Python 测试（pytest）；kernel/crates/*/tests/ Rust 测试 |
scripts/ |
运维/清理脚本 |
data/、logs/、reports/ |
运行时产物 |
存储：SQLite（默认项目根 agentos_kernel.db），driver 化切换见
config/storage.yaml（AGENTOS_STORAGE_DRIVER/AGENTOS_DB_PATH 环境变量可覆盖）。
⚠️ 工作区铁律（最重要）
受管开发环境（本仓容器工作区）会被周期性还原到 git HEAD——未提交的改动会被抹掉
（普通 clone 无此行为，但提交纪律通用）。任何修改（含中间产物、测试数据）完成后立即
commit，不要留到"最后一起提交"。commit 前的调查/验证工作尽量用未跟踪文件（??
状态）或文档目录进行。
架构要点（2026-08 现状）
- 一切皆插件（架构公理，写文档/方案一律以此为基）：内核只是执行基座（管道解释
执行/能力注册表/插件装载/存储），一切能力——LLM、记忆、评估、审批、触发器、通道、
主题、乃至 agent 配置的加载——皆插件承载；改业务行为 = 加/改插件或配置，不动内核。
动因六条（自进化落地/改动半径分离/故障隔离/语言生态自由/统一契约/治理单点）与
能力→插件对照表见
docs/ARCHITECTURE.md设计哲学 §1。 - 插件即声明：插件 =
plugin.json清单 + Python 实现。capabilities.tools= LLM 工具声明即注册（无需类型豁免）；capabilities.services= 内部服务方法元数据。 工具声明要带output_schema+render（工具契约，tool_core 校验 fail-closed， 前端按 render 意图路由）。插件放plugins/shared/{system,tools,pipeline}/<name>/。 - 评估闸门：插件判定，内核只落库：判定逻辑已从内核移出归插件——评估裁决
在
plugins/shared/system/evaluation/（task_evaluate 工具），放行检测在管道 output 步骤 task_reminder（提醒耗尽仍无评估证据 →task.status = pending_evaluation，不落 completed；有证据内核才补落默认 completed）；内核不 做判定，只经 pipeline-state 写面记录结果。agent 配置加载同样已移出内核，由 管道输入插件context_build自持（plugins/shared/pipeline/input/context_build/）。 - 任务默认隔离执行：默认工作空间
workspace/{task_id}+ isolated。 - 工具面过滤：LLM 可见工具由
config/agents/main/agentos.yaml（及executor/general_agent.yaml）的tool_ids白名单控制，新工具记得加入。 - 权限模式：5 种权限模式 + 参数级危险判定，纯插件前端（http_endpoints + form compact + human-interaction 确认）。会话隔离由 isolation_guard 容器落地。
- 插件热发现/热重载全链路：新建插件目录、修改 plugin.json、改插件 Python 代码 均由 watcher 自动处理（发现→G2 校验→注册/重注册/respawn），无需 re-enable 或重启； cdylib 集合变更走 G8 自动重启（同 id 换产物保守重启）；已知插件面取自共享 manifests 活集合（新插件热注册后管道引用即可编译）；前端 schema 需刷新页面才更新。
- 多循环体/执行上下文：
execution_context贯穿任务链；agent_id 的本质是执行上下文 键——内核无 Agent 运行时对象，语义由插件按键展开（context_build 按键加载配置， 工具面tool_ids同属这份执行上下文，与提示词同源注入）；内核零 agent 配置 知识（不读config/agents/**）——工具面过滤 = tool_schema 插件读 state.tool_ids 经内核tool-surfacecapability 服务完成（ADR 2026-09-02）。
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 · 141 lines · 3,156 tokens per session scan A 3ed0443f8bee
Agent-os-open AGENTS.md is an instructions file published in the GitHub repository jianchen08/Agent-os-open (5 stars, last pushed 2d ago), licensed Apache-2.0. It adds 3,156 tokens to every session, about $0.0158 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 2026-09-05.
Other instructions, from other repositories
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
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).
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.