How real projects brief Codex, Cursor and every other agent that reads AGENTS.md.
pizza-bot-appAGENTS.mdpizza-bot-app/pizza-bot/AGENTS.md
runtime (DeepAgents/LangGraph). The runtime emits native `@langchain/protocol` `ProtocolEvent` frames via `streamEvents(v3)`; every frontend consumes them through the `@langchain/langgraph-sdk` `Client`/`ThreadStream`/`StreamController` over HTTP/SSE. The seam is that **protocol
devforthAGENTS.mddevforth/adminforth/AGENTS.md
inline regex literals when reused or non-trivial --- ## Trust the contract If backend and frontend are part of the same system, and the backend explicitly guarantees a response shape
GRCEngClubAGENTS.mdGRCEngClub/claude-grc-engineering/AGENTS.md
Node CLI scripts, the test suite, the optional local dashboard server, and/or the Vite frontends. ## Setup, tests, and checks - Runtime: Node.js (CI pins Node 20; Node 22 works). Single
ergreletAGENTS.mdergrelet/windiff/AGENTS.md
downloading PE binaries from Winbindex and extracting symbol/type/syscall data from matching PDBs. - **windiff_frontend** — Next.js/React static site that loads those databases and provides Browse/Diff UI. ## Commands ### windiff_cli (Rust
wesammustafaAGENTS.mdwesammustafa/opencode-primer/docs/agents.md
description, mode, and permissions), or just write the markdown by hand. ```markdown --- description: Senior frontend engineer — Tailwind + React + TypeScript expert mode: all model: anthropic/claude-sonnet-5 temperature: 0.2 permission: edit: allow bash
labsaiAGENTS.mdlabsai/EDDI/AGENTS.md
built in `prepare-package` and copied into the jar just before it is assembled (`frontend-maven-plugin` + `maven-resources-plugin` in `pom.xml`). So `compile`, `test` and `quarkus:dev` never
smart-mcp-proxyAGENTS.mdsmart-mcp-proxy/mcpproxy-go/AGENTS.md
upstream orchestration live in `internal/` (`internal/runtime`, `internal/httpapi`, `internal/upstream`, `internal/storage`). - Vue/Tailwind assets are kept in `frontend/`; bundled static output is embedded from `web/`. - Integration and regression assets reside under `tests
yan5xuAGENTS.mdyan5xu/codexloom/AGENTS.md
does it only defend against an attacker who already controls the local Owner account? ## Frontend package manager The WebUI uses pnpm 11.1.2, pinned by the `packageManager` field in `web/package.json
delexwAGENTS.mddelexw/claude-code-trace/AGENTS.md
format cargo fmt --manifest-path src-tauri/Cargo.toml # Rust format # Test npx vitest run # Frontend tests npm run test:e2e # Playwright e2e: real headless backend + real browser (see e2e/) cargo
xiaojilele-glitchAGENTS.mdxiaojilele-glitch/WhyBuddy/AGENTS.md
commands are in `package.json` scripts: - `pnpm run dev:all` — start the full stack: Vite frontend `:3000`, Node server `:3001`, Lobster executor `:3031`, and Python engine `:9700`. This is the normal
autonomous-aiAGENTS.mdautonomous-ai/autonomous-os/AGENTS.md
lint # Catch broken local imports + undefined names make hal-test # Run HAL tests # Web frontend (React/Vite/Tailwind in system/web) make web-install # npm install make web-dev # Vite dev server make
monarchjunoAGENTS.mdmonarchjuno/tradingcodex/AGENTS.md
Keep durable service behavior under `tradingcodex_service/application/` and use direct canonical imports. - Frontend source lives under `frontend/`; compiled assets under `tradingcodex_service/static/tradingcodex_web/` are generated, not hand-edited. - Keep prompts, skills
timwuhaotianAGENTS.mdtimwuhaotian/the-pair/AGENTS.md
humans to observe and intervene. ## Tech Stack - **Framework:** Tauri 2.0 - **Backend:** Rust - **Frontend:** React 19, TypeScript - **Styling:** Tailwind CSS v4, `clsx`, `tailwind-merge` - **Icons:** `lucide-react` - **State Management:** `zustand` (`usePairStore
SentryAGENTS.mdgetsentry/junior/AGENTS.md
Repo-wide policy index | `policies/README.md` | | Runtime vocabulary | `TERMINOLOGY.md` | | Design and failures | `policies/interface-design.md`, `policies/correctness-complexity.md`, `policies/error-handling.md` | | Frontend components | `policies/frontend-components.md`, `packages/junior-dashboard/src/client/components/README.md` | | Agent steering | `policies/agent-steering.md` | | Provider boundaries | `policies/provider-boundaries.md` | | Comments and telemetry | `policies/code-comments.md`, `policies/observability.md`, `TELEMETRY.md
lovstudioAGENTS.mdlovstudio/Ataru/AGENTS.md
上下文；Run 结果才可以使用用户首句作为标题。禁止为 Turn 或 Run 使用 `Untitled turn`、`未命名回合` 之类的标题占位。 ## Commands ```bash # Frontend development (hot reload) pnpm dev # Type check + production build pnpm build # Run Tauri desktop
milady-aiAGENTS.mdmilady-ai/milady/AGENTS.md
Electrobun), web dashboard, and platform connectors (Telegram, Discord, WeChat, etc.). Inherits `~/.codex/AGENTS.md`. ## Cloud frontend visual review (REQUIRED for any UI change) Any change in `eliza/packages/cloud-frontend/` MUST go through the manual
spencermarxAGENTS.mdspencermarx/open-code-review/openspec/AGENTS.md
tasks.md:** ```markdown ## 1. Implementation - [ ] 1.1 Create database schema - [ ] 1.2 Implement API endpoint - [ ] 1.3 Add frontend component - [ ] 1.4 Write tests ``` 5. **Create design.md when needed:** Create `design.md
plmbrAGENTS.mdplmbr/notebook-intelligence/AGENTS.md
repo's Prettier (this covers Markdown too). ## JavaScript tooling - Use `jlpm` for every frontend command (install, build, lint, test). Do not use a system `yarn` or `npm`. `jlpm
earthtojakeAGENTS.mdearthtojake/step.parts/AGENTS.md
recolor models for aesthetics. - Verify both directory cards and detail pages. For frontend work, avoid landing-page-style redesigns. This is a utilitarian catalog: dense, searchable, and scan-friendly. Cards
outsourc-eAGENTS.mdoutsourc-e/clawsuite/AGENTS.md
electron, ipc, main process, preload | `aurora-electron` | | anything else | `aurora-coder` (default) | --- ## aurora-coder — Frontend Specialist **Role:** React/TypeScript UI work. Owns all workspace screens and components. **Model:** codex
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.
