AGENTS.md examples from real projects

chakra-uiAGENTS.md

panda

chakra-ui/panda/AGENTS.md

/vite/ # @pandacss/vite — bundler integration /postcss/ # @pandacss/postcss — PostCSS plugin /preset-base/ # @pandacss/preset-base /preset-panda/ # @pandacss/preset-panda /mcp/ # @pandacss/mcp — MCP server on the compiler driver /compiler/ # @pandacss/compiler — NAPI wrapper around Rust engine /crate/ # compiler_napi

bethingtonAGENTS.md

ghidra-mcp

bethington/ghidra-mcp/AGENTS.md

documentation workflows ## Directory Structure - `src/` — Java source for Ghidra extension and headless server - `python/bridge_mcp_ghidra/` — Python MCP bridge package (`bridge-mcp-ghidra` console script / `python -m bridge_mcp_ghidra`) - `pyproject.toml` + `uv.lock

lightdashReads credentialsAGENTS.md

lightdash

lightdash/lightdash/AGENTS.md

server. Chart-as-code JSON schema is generated from backend OpenAPI: ```bash pnpm generate:chart-as-code-schema pnpm check:chart-as-code-schema ``` **MCP Tool Snapshot:** The committed `packages/common/src/schemas/json/mcp-tools-1.0.json

No other project uses it yet

lemonade-sdkReads credentialsAGENTS.md

lemonade

lemonade-sdk/lemonade/AGENTS.md

drift. | | `src/cpp/server/anthropic_api.cpp` | Anthropic API compatibility | | `src/cpp/server/ollama_api.cpp` | Ollama API compatibility | | `src/cpp/server/mcp_server.cpp` | MCP gateway (POST /mcp) | | `src/cpp/include/lemon/websocket_server.h` | WebSocket Realtime API server | | `src/cpp/include/lemon/model_types.h` | Model type and device type enums | | `src/cpp/include/lemon/config_file.h` | config.json load/save/migrate | | `src/cpp/include/lemon/recipe_options.h

No other project uses it yet

mostlygeekAGENTS.md

llama-swap

mostlygeek/llama-swap/AGENTS.md

Docs Agent against a local model after changing its system prompt, `docs/kb/` content, the MCP tool descriptions, or the docs search ranking. See `evals/docs-agent/README.md` for the tuning loop. ## Documentation - When

No other project uses it yet

netease-youdaoAGENTS.md

LobsterAI

netease-youdao/LobsterAI/AGENTS.md

agent experience that can work with local projects, files, browser previews, IM channels, skills, MCP servers, scheduled tasks, and rich artifacts. ### Cowork vs OpenClaw `Cowork` is LobsterAI's product/session layer

foxcppAGENTS.md

maddy

foxcpp/maddy/AGENTS.md

AGENTS.md — Maddy Mail Server ## Architecture Maddy is a composable all-in-one mail server (MTA/MX/IMAP) written in Go. The core abstraction is the **module system**: every functional component (auth, storage

No other project uses it yet

soopersetReads credentialsAGENTS.md

mcp-atlassian

sooperset/mcp-atlassian/AGENTS.md

mixins (pages, search, analytics, …) | | ` ├─ models/` | Pydantic v2 data models (`ApiModel` base) | | ` ├─ servers/` | FastMCP server instances (`jira_mcp`, `confluence_mcp`) | | ` ├─ preprocessing/` | Content conversion (ADF/Storage → Markdown) | | ` └─ utils/` | Shared utilities (auth, logging

No other project uses it yet

dograh-hqAGENTS.md

dograh / api

dograh-hq/dograh/api/AGENTS.md

data access ├── schemas/ # Pydantic request/response schemas ├── tasks/ # Background jobs and post-call work ├── mcp_server/ # MCP surface exposed by the backend ├── utils/ # Shared utilities ├── alembic/ # Database migrations └── tests/ # Test suite

likec4AGENTS.md

likec4

likec4/likec4/AGENTS.md

tool for visualizing software architecture. It provides a DSL for describing architecture, a language server, CLI, VSCode extension, and web-based diagram visualization. ## Read task guidance before work AGENTS.md

Microsoft AzureAGENTS.md

azure-sdk-for-net

Azure/azure-sdk-for-net/AGENTS.md

code - Useful for debugging Azure.Core and other dependencies ## Agent-Specific Tools and MCP ### MCP Server Requirements To use MCP (Model Context Protocol) tool calls: - **PowerShell must be installed** ([Installation Guide

CloudflareReads credentialsAGENTS.md

agents

cloudflare/agents/AGENTS.md

Main showcase app — all SDK features in one UI (uses Kumo design system) mcp/ # MCP server example mcp-client/ # MCP client example ... # ~20 examples total experimental/ # Work-in-progress experiments

No other project uses it yet

CloudflareReads credentialsAGENTS.md

agents / examples

cloudflare/agents/examples/AGENTS.md

simple, clear, and consistent. Each example should focus on **one feature or concept** (e.g., MCP servers, email routing, workflows). The exception is `playground/`, which is the kitchen-sink showcase covering

No other project uses it yet

CloudflareReads credentialsAGENTS.md

agents / agents

cloudflare/agents/packages/agents/AGENTS.md

wrapper for SDK v2 stateless servers | | `agents/mcp/client` | `src/mcp/client/index.ts` | MCP client manager (connect to remote MCP servers from an Agent) | | `agents/email` | `src/email.ts` | Email routing, resolvers, header signing | | `agents/workflows` | `src/workflows.ts` | `AgentWorkflow` — Workflows

No other project uses it yet

agentscope-aiAGENTS.md

HiClaw

agentscope-ai/HiClaw/AGENTS.md

tests/) — integration test suite ### To modify Higress routing and initialization - [manager/scripts/init/setup-higress.sh](manager/scripts/init/setup-higress.sh) — route, consumer, MCP server setup (local / full console mode) - [design/higress-console-api.yaml](design/higress-console-api.yaml) — Higress Console API spec (OpenAPI 3.0) ## Technology

VercelAGENTS.md

eve

vercel/eve/AGENTS.md

answer back. 5. **Discoverability determines whether agents find you at all.** `llms.txt`, typed SDKs, MCP servers, machine-readable specs. If an agent can't discover your surface, it reaches

VercelReads credentialsAGENTS.md

eve / eve-software-factory-template

vercel/eve/apps/templates/eve-software-factory-template/AGENTS.md

brain`, gated by `factoryBrainPolicy` so unattended runs can't write shared context). - **Connections** are MCP servers: `linear` (`https://mcp.linear.app/mcp`, app-scoped auth shared through `linearAuth` in `agent/lib/constants.ts`). Its connection

VercelReads credentialsAGENTS.md

eve / kody-eve-template

vercel/eve/apps/templates/kody-eve-template/AGENTS.md

ctx.agent()` and requests structured output per turn (`defineAgent` no longer takes `outputSchema`). - **Connections** are MCP servers: `linear` (`https://mcp.linear.app/mcp`, app-scoped auth shared through `linearAuth` in `agent/lib/constants.ts` with scopes

VercelReads credentialsAGENTS.md

eve / marketing-team-eve-template

vercel/eve/apps/templates/marketing-team-eve-template/AGENTS.md

Mail cannot be recalled, so the Resend surface is an allow list.** Resend's MCP server publishes around 85 tools spanning campaigns, contacts, domains, webhooks, and API keys, of which

awarexoneAGENTS.md

Agentic-Bug-Hunter

awarexone/Agentic-Bug-Hunter/AGENTS.md

poisoning | | `skills/graphql-audit/` | GraphQL hunting — introspection, field suggestions, batching DoS, IDOR via aliasing, injection | | `skills/mcp-server-audit/` | MCP server audit — tool poisoning, param→sink injection, missing approval gates, secret leaks, rug-pull/confused-deputy