How real projects brief Codex, Cursor and every other agent that reads AGENTS.md.
0xStephAGENTS.md0xSteph/pentest-ai/AGENTS.md
every commit. ## Layout | Path | Purpose | |---|---| | `cli/` | `ptai` CLI entrypoint, menu, install wizard, MCP setup | | `mcp_server/` | FastMCP server, exposes tools to Claude Code, Cursor, etc. | | `api/` | FastAPI HTTP surface (alternative
shiwenwenAGENTS.mdshiwenwen/hope-agent/AGENTS.md
EventBus`，内核不用 `APP_HANDLE`。所有壳在 `init_runtime` 前调用 `ha_server::wire_features()`；新增全局装配特征只修改该入口和 `ha-server/Cargo.toml`，其它壳直接调用该特征 API 时才添加依赖。 - 配置读 `cached_config()`、写 `mutate_config((category, source), …)`，禁止克隆后自行加载、修改、保存
kubeshopAGENTS.mdkubeshop/testkube/AGENTS.md
cmd/tools` contains internal tooling for release management and version bumping. ## MCP integration - `pkg/mcp/` implements the Model Context Protocol server for AI assistant integration. - Exposes tools across workflows, executions, artifacts
JetBrainsAGENTS.mdJetBrains/MPS/AGENTS.md
available in your session, ask the user to start MPS and enable the MPS MCP server before continuing with any MPS work. - Do not attempt to parse, edit, or reason
agentlas-aiAGENTS.mdagentlas-ai/Agentlas-OS/AGENTS.md
APIs, model providers, hosted browsers, data sources, SaaS connectors, payment providers, and vendor-specific MCP servers. Reject a provider contribution when it requires provider-specific changes to Core registries
KaelioAGENTS.mdKaelio/ktx/AGENTS.md
this repository (Codex, Claude, Gemini, and similar tools). Do not assume an external app server, frontend, database migrations, ORPC contracts, or `python-service/` layout exist here. ## Critical Rules ### Absolute Requirements
fynnflueggeAGENTS.mdfynnfluegge/agtx/AGENTS.md
handling) ├── db/schema.rs # SQLite operations (tasks, projects, notifications) ├── db/models.rs # Task, Project, TaskStatus structs ├── mcp/server.rs # MCP server (JSON-RPC over stdio) ├── tmux/ # Tmux session/window management ├── git/ # Worktree creation, conflict detection ├── agent/ # Agent
silversteinAGENTS.mdsilverstein/minutes/AGENTS.md
toolchain + UI items below. **Read it before any commit that touches Rust, MCP server, frontend, or release surfaces** — the table is the single source of truth and is kept
limecloudAGENTS.mdlimecloud/lime/AGENTS.md
持久化与工具 owner，不建立第二套业务后端。未来 Cloud 只能通过认证 transport 接入同一 App Server 边界，不预建平行 runtime。 5. Agent loop、状态机、Thread/Turn/Item、工具生命周期、MCP、Skills、Multi-Agent、历史恢复和 GUI 护栏对齐 `/Users/coso/Documents/dev/rust/codex`。多模型控制面的 catalog、默认选择
Model Context ProtocolAGENTS.mdmodelcontextprotocol/kotlin-sdk/AGENTS.md
kotlin-sdk-client Client: MCP client implementation (tools, prompts, resources, sampling) kotlin-sdk-server Server: MCP server implementation (feature registries, sessions) kotlin-sdk-testing ChannelTransport for in-memory client-server
inkeepAGENTS.mdinkeep/agents/AGENTS.md
Update baselines after intentional UI changes pnpm --filter @inkeep/agents-manage-ui test:visual:update # Stop the server when done docker compose -f docker-compose.visual.yml down ``` - Visual test files: `*.browser.test.tsx` - Baselines stored in `src/__screenshots__
chunkhoundAGENTS.mdchunkhound/chunkhound/AGENTS.md
into searchable knowledge bases for AI assistants ## MODIFICATION_RULES **NEVER:** - NEVER Use print() in MCP server (stdio.py, http_server.py, tools.py) - NEVER Make single-row DB inserts in loops - NEVER Use forward
Hugging FaceAGENTS.mdhuggingface/Mongoku/AGENTS.md
able to use the Svelte MCP server, where you have access to comprehensive Svelte 5 and SvelteKit documentation. Here's how to use the available tools effectively: ## Available MCP Tools
yohey-wAGENTS.mdyohey-w/multi-agent-shogun/AGENTS.md
Codex CLI + tmux multi-agent parallel dev platform with sengoku military hierarchy
preset-ioAGENTS.mdpreset-io/agor/AGENTS.md
through a delegated external substrate. | | **MCP** | Model Context Protocol. Agor exposes itself as an MCP server (`POST /mcp`) so agents can introspect sessions, branches, boards, etc. | | **RBAC** | Always-on normalized
m0n0x41dAGENTS.mdm0n0x41d/haft/AGENTS.md
coding agent (Claude Code, Codex, OpenCode, Cursor), this CLI for manual access, and the MCP server (`haft serve`) for programmatic access. Standalone interactive agent (`haft agent`), TUI, and desktop wrappers
miscusi-peekAGENTS.mdmiscusi-peek/cheatengine-mcp-bridge/AGENTS.md
Table Lua Script`; in that case execute `dofile([[C:\path\to\cheatengine-mcp-bridge\MCP_Server\ce_mcp_bridge.lua]])` instead of pasting the full bridge. Success log: `[MCP v12.0.0] MCP Server Listening
thetahealthAGENTS.mdthetahealth/mirobody/AGENTS.md
storage + MCP | `mirobody/documents/`, `collect/`, `utils/`, `user/`, `task/`, `mcp/` | `[parse]` / `[app]` | no `langchain*`, `langgraph`, `deepagents` | | ③ Agent | `mirobody/agent/` (one agent: `MirobodyAgent`, on `deepagents`), `server/` | `[agent]` (the harness as a library) / `[app]` | anything
sceneviewAGENTS.mdsceneview/sceneview/AGENTS.md
react-native/` | Native bridges (Android + iOS) | | `samples/` | One demo app per platform | | `mcp/` | `sceneview-mcp` server + packages | Design tokens live in `DESIGN.md` — read it before generating any UI, never hardcode
puppyone-aiAGENTS.mdpuppyone-ai/puppyone-cloud/AGENTS.md
short links) │ ├── internal/ # Internal API (X-Internal-Secret) │ └── utils/ # Utilities (logging/middleware) ├── mcp_service/ # Standalone MCP Server service (FastMCP) ├── sql/ # Database DDL & migrations ├── tests/ # Tests ├── scripts/ # Scripts └── docs/ # Feature documentation ``` ### Development
An open format for instructions to coding agents, read by Codex, Cursor and others. Think of it as a README written for agents.
At the repository root, with more specific files in subdirectories. Agents read the one closest to the file they're editing.
Setup and test commands, code style, and the rules a new contributor would need to know.
Claude Code reads CLAUDE.md. A one-line CLAUDE.md that points at AGENTS.md covers both.
