sdk-python / strands-py
strands-agents/sdk-python/strands-py/AGENTS.md
while the SDK was *running* that an operator would want in their logs (an MCP server failed to start, a tool failed to load, a store write failed). These
How real projects brief Codex, Cursor and every other agent that reads AGENTS.md.
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
strands-agents/sdk-python/strands-py/AGENTS.md
while the SDK was *running* that an operator would want in their logs (an MCP server failed to start, a tool failed to load, a store write failed). These
strands-agents/harness-sdk/strands-py/AGENTS.md
while the SDK was *running* that an operator would want in their logs (an MCP server failed to start, a tool failed to load, a store write failed). These
Steady34 mentions in 2 weeks
No other project uses it yet
apache/shenyu/AGENTS.md
dubbo, grpc, spring-cloud, sofa, tars, websocket, mqtt...), `-security`, `-logging`, `-cache`, `-fault-tolerance`, `-ai`, `-mcp-server`, etc. A plugin is activated by adding its starter to the bootstrap pom. - **`shenyu
rikkahub/rikkahub/AGENTS.md
conversation is persisted it also holds a `ConversationConfig` snapshot (chat model, reasoning level, search, MCP servers, workspace, skills) taken from the assistant; from then on chat-page changes to those
No other project uses it yet
ever-co/ever-gauzy/AGENTS.md
affected`) instead of using the underlying tooling directly - You have access to the Nx MCP server and its tools, use them to help the user - When answering questions about
Gentleman-Programming/gentle-ai/docs/agents.md
configuration, so Gentle AI makes no Conductor-specific writes: it does not install skills, MCP servers, or system prompt files for Conductor — configure those through Claude Code instead
storytold/effectcraft/AGENTS.md
crates/engine`): id, label, menu path, shortcut, params, enabled(), run(). UI, CLI, control channel and MCP dispatch by id. - **Everything is agent-drivable:** every interactive widget registers an automation
Steady17 mentions in 2 weeks
No other project uses it yet
storytold/effectcraft/docs/agents.md
macOS with the CLI unzipped onto PATH claude mcp add effectcraft -- effectcraft-cli mcp ``` The server speaks JSON-RPC 2.0 over stdio, one message per line, and supports MCP protocol
Steady17 mentions in 2 weeks
No other project uses it yet
meshtastic/firmware/AGENTS.md
embedded codebase targeting ESP32 / nRF52 / RP2040 / STM32WL / Linux-Portduino LoRa mesh radios. The Python MCP server that AI agents use to flash, configure, and test connected devices now lives
agegr/pi-web/AGENTS.md
redirected address or code (Models, MCP) ProjectTrustDialog.tsx trust confirmation listing the project's MCP servers AgentsConfig.tsx built-in subagent toggle + agent profile editor SessionChatPanel.tsx right-panel tab of a subagent
webiny/webiny-js/AGENTS.md
publish with refreshed references ## Webiny This project uses the Webiny framework. A `webiny` MCP server is available. When helping with Webiny-related tasks: 1. Call `list_webiny_skills
alphaXiv/OpenResearch/arc/agents.md
worktree. **orx never edits the user's agent configuration.** What it adds (hooks, an MCP server, sandbox roots) goes through the agent's command line or a settings file
Steady8 mentions in 2 weeks
No other project uses it yet
ludovic111/horizon/AGENTS.md
named harnesses python3 tools/lan_e2e.py --check chest # two real games over LAN (server-thread host) python3 tools/agent_control_e2e.py # horizon-ctl / MCP end to end ``` - Harnesses and the game need the vanilla data
No other project uses it yet
openai/plugins/plugins/data-analytics/AGENTS.md
widget and portable-reader bundles, `npm run typecheck` for TypeScript, and `npm test` for MCP server, reader, builder, and widget contract coverage. Reader changes must also pass `npm run test
samuelclay/NewsBlur/AGENTS.md
Other useful endpoints: `vitals.crashrate`, `vitals.anrrate`, `vitals.errors.counts`. ## Browser Testing - **Do NOT use the Chrome DevTools MCP server unless explicitly asked** — the user will verify manually - Local dev: `https://localhost` for main
superdesigndev/treg/AGENTS.md
proxy that makes the caller's **real upstream request**, injects the credential server-side when one is required, and relays the answer verbatim. A catalog endpoint explicitly verified as public
4gray/iptvnator/AGENTS.md
project tasks through local `pnpm nx`, not a global Nx installation. - Use the Nx MCP server when available; otherwise use CLI discovery. - Check `node_modules/@nx/ /PLUGIN.md` for plugin guidance
Sequel-Ace/Sequel-Ace/AGENTS.md
file (e.g. `SAFavoriteItem.swift` vs `SAFavoriteItem+Tree.swift`). ## Xcode automation (MCP) - The repo ships an `xcode` MCP server in `.mcp.json`. It runs `xcrun mcpbridge`, which is bundled with Xcode 26+ — no install
Tencent/AI-Infra-Guard/AGENTS.md
agent ./cmd/agent AIG_SERVER=127.0.0.1:8088 ./agent ``` ### Docker 启动 ```bash # 使用预构建镜像 docker-compose -f docker-compose.images.yml up -d # 使用本地源码构建镜像 docker-compose up -d ``` ### Python 子模块 ```bash # MCP 扫描 pip install
Steady2 mentions in 2 weeks
No other project uses it yet
Devin-AXIS/A2V/AGENTS.md
AGENTS.md iPolloWork helps users run agents, skills, and MCP. It is a source-available alternative to Claude Cowork/Codex as a desktop app. ## What iPolloWork Is iPolloWork is a practical control
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.