Skip to content
Back to search

world2agent

World2Agent(W2A) is an open protocol that standardizes how Al agents perceive the real world.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked yesterday
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/machinepulse-ai--world2agent/badge.svg)](/listings/machinepulse-ai--world2agent)

HTML

<a href="/listings/machinepulse-ai--world2agent"><img src="/api/listings/machinepulse-ai--world2agent/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

W2A ships with native [agent-runtime plugins](https://github.com/machinepulse-ai/world2agent-plugins) for Claude Code, Hermes, and OpenClaw — pick whichever runtime you already use.

Claude Code

In an active session, install the `world2agent` plugin:

/plugin marketplace add machinepulse-ai/world2agent-plugins
/plugin install world2agent@world2agent-plugins
/reload-plugins

Add a sensor — for example, Hacker News stories, frontier AI lab posts:

/world2agent:sensor-add @world2agent/sensor-hackernews
/world2agent:sensor-add @quill-io/sensor-frontier-ai-news

Restart Claude Code with the plugin channel loaded so sensor signals flow into your session:

claude --dangerously-load-development-channels plugin:world2agent@world2agent-plugins

Hermes

Install the bridge once:

npm install -g @world2agent/hermes-sensor-bridge
hermes skills install machinepulse-ai/world2agent-plugins/hermes-sensor-bridge/skills/world2agent-manage

In an interactive `hermes` session, describe the intent in natural language or use the slash form — the agent handles npm install, `SETUP.md` Q&A, webhook subscription, and subprocess startup:

/world2agent-manage add @world2agent/sensor-hackernews

> First time only: the agent will ask you to restart `hermes gateway` once after it enables the webhook platform. Every install after that is seamless.

Each signal triggers a fresh `AIAgent.run_conversation()` against the generated handler skill.

OpenClaw

Three steps:

npm install -g @world2agent/openclaw-sensor-bridge
openclaw skills install world2agent-manage

Then send this in your OpenClaw chat:

Use world2agent-manage skill install @quill-io/sensor-frontier-ai-news

The skill walks the SETUP.md Q&A, generates a handler skill, registers the sensor, and starts the supervisor. Subsequent signals each trigger a fresh `/hooks/agent` call against the handler.

> First time only: the bridge writes a managed `hooks` block into `~/.openclaw/openclaw.json` (auto-generates `hooks.token` if absent) and asks you to run `openclaw gateway restart` once. A timestamped backup of the original config is kept next to the file. Every install after that is seamless.

If you have a paired chat platform (Feishu, iMessage, Telegram, …) configured via `<PLATFORM>_HOME_CHANNEL` in `~/.openclaw/.env`, replies are auto-pushed to that chat by default.

---

→ Browse the full catalog on [SensorHub](https://world2agent.ai/hub).

> **Security — install only sensors you trust.** A sensor's signals drive what your agent perceives and does, so an untrusted sensor is effectively an untrusted instruction source. Stick to open-source sensors from authors you trust, and review the code first.

**Integrating W2A into your own agent system?** See the [developer quick start](./docs/quick-start.md#option-2-code--sdk--sensor) for the SDK code path.

→ [Full guide](./docs/quick-start.md) · [Multi-sensor](./docs/multi-sensor.md) · [SensorHub](./docs/sensorhub.md)

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 193 lines
  1. ![Welcome to World2Agent](https://world2agent.ai/assets/readme-banner.png)
  2. <p align="center">
  3. <i>Agents can't act on what they can't perceive.</i>
  4. </p>
  5. <p align="center">
  6. <a href="./LICENSE"><img src="https://img.shields.io/badge/License-Apache%202.0-blue.svg" alt="License" /></a>
  7. <a href="https://www.npmjs.com/org/world2agent"><img src="https://img.shields.io/badge/npm-%40world2agent-red" alt="npm" /></a>
  8. </p>
  9. <p align="center">
  10. <a href="https://world2agent.ai">Website</a> ·
  11. <a href="#quick-start">Quick Start</a> ·
  12. <a href="#sensors">Sensors</a> ·
  13. <a href="https://world2agent.ai/hub">SensorHub</a> ·
  14. <a href="./docs">Docs</a> ·
  15. <a href="#community">Community</a>
  16. </p>
  17. <!-- Concept Video -->
  18. <p align="center">
  19. <a href="https://world2agent.ai/assets/promo-w2a.mp4">
  20. Watch the W2A Concept Video
  21. </a>
  22. </p>
  23. <p align="center">
  24. <img src="https://world2agent.ai/assets/github-star.gif" width="600" alt="star" />
  25. </p>
  26. ***
  27. ## What is World2Agent?
  28. World2Agent (W2A) is an open protocol that standardizes how AI agents perceive the real world. Install a sensor, your agent gets structured, real-time data. Swap sensors freely — they all speak the same schema.
  29. W2A isn't a product. It's an open protocol and an invitation. We built the first sensors — the real breakthroughs will come from the community.
  30. → [Why W2A? Full story](./docs/why-w2a.md)
  31. ## Partners
  32. When using the W2A, we welcome you to use our partner Zeabur to deploy your code or agents, allowing users worldwide to experience them online!
  33. <p align="center">
  34. <a href="https://zeabur.com" target="_blank">
  35. <img src="https://world2agent.ai/assets/zeabur-logo.svg" width="150">
  36. </a>
  37. </p>
  38. ## Architecture
  39. **World → Sensor → Agent**
  40. Sensors watch data sources and emit structured data following W2A Protocol. Your agent receives signals and decides what to do.
  41. ![World2Agent system architecture](./docs/images/system-architecture.png)
  42. → [Signal format spec](./docs/signal-format.md) · [Architecture deep dive](./docs/architecture.md)
  43. ## Quick Start
  44. W2A ships with native [agent-runtime plugins](https://github.com/machinepulse-ai/world2agent-plugins) for Claude Code, Hermes, and OpenClaw — pick whichever runtime you already use.
  45. ### Claude Code
  46. In an active session, install the `world2agent` plugin:
  47. ```
  48. /plugin marketplace add machinepulse-ai/world2agent-plugins
  49. /plugin install world2agent@world2agent-plugins
  50. /reload-plugins
  51. ```
  52. Add a sensor — for example, Hacker News stories, frontier AI lab posts:
  53. ```
  54. /world2agent:sensor-add @world2agent/sensor-hackernews
  55. /world2agent:sensor-add @quill-io/sensor-frontier-ai-news
  56. ```
  57. Restart Claude Code with the plugin channel loaded so sensor signals flow into your session:
  58. ```bash
  59. claude --dangerously-load-development-channels plugin:world2agent@world2agent-plugins
  60. ```
  61. ### Hermes
  62. Install the bridge once:
  63. ```bash
  64. npm install -g @world2agent/hermes-sensor-bridge
  65. hermes skills install machinepulse-ai/world2agent-plugins/hermes-sensor-bridge/skills/world2agent-manage
  66. ```
  67. In an interactive `hermes` session, describe the intent in natural language or use the slash form — the agent handles npm install, `SETUP.md` Q&A, webhook subscription, and subprocess startup:
  68. ```
  69. /world2agent-manage add @world2agent/sensor-hackernews
  70. ```
  71. > First time only: the agent will ask you to restart `hermes gateway` once after it enables the webhook platform. Every install after that is seamless.
  72. Each signal triggers a fresh `AIAgent.run_conversation()` against the generated handler skill.
  73. ### OpenClaw
  74. Three steps:
  75. ```bash
  76. npm install -g @world2agent/openclaw-sensor-bridge
  77. openclaw skills install world2agent-manage
  78. ```
  79. Then send this in your OpenClaw chat:
  80. ```
  81. Use world2agent-manage skill install @quill-io/sensor-frontier-ai-news
  82. ```
  83. The skill walks the SETUP.md Q&A, generates a handler skill, registers the sensor, and starts the supervisor. Subsequent signals each trigger a fresh `/hooks/agent` call against the handler.
  84. > First time only: the bridge writes a managed `hooks` block into `~/.openclaw/openclaw.json` (auto-generates `hooks.token` if absent) and asks you to run `openclaw gateway restart` once. A timestamped backup of the original config is kept next to the file. Every install after that is seamless.
  85. If you have a paired chat platform (Feishu, iMessage, Telegram, …) configured via `<PLATFORM>_HOME_CHANNEL` in `~/.openclaw/.env`, replies are auto-pushed to that chat by default.
  86. ---
  87. → Browse the full catalog on [SensorHub](https://world2agent.ai/hub).
  88. > **Security — install only sensors you trust.** A sensor's signals drive what your agent perceives and does, so an untrusted sensor is effectively an untrusted instruction source. Stick to open-source sensors from authors you trust, and review the code first.
  89. **Integrating W2A into your own agent system?** See the [developer quick start](./docs/quick-start.md#option-2-code--sdk--sensor) for the SDK code path.
  90. → [Full guide](./docs/quick-start.md) · [Multi-sensor](./docs/multi-sensor.md) · [SensorHub](./docs/sensorhub.md)
  91. ## Sensors
  92. ### [SensorHub](https://world2agent.ai/hub)
  93. **SensorHub is the catalog of every W2A sensor** — official and community-built, organized by what each one perceives (markets, news, production alerts, weather, AI labs, …). Browse it to see what's available, view each sensor's signal schema, and grab the one-line install command. **Looking for a sensor? Start here.**
  94. Every sensor is a standard npm package — `npm search w2a-sensor` works as a fallback if you prefer the raw view.
  95. → [SensorHub guide](./docs/sensorhub.md)
  96. ### Missing a sensor?
  97. [Build your own](./docs/build-a-sensor.md) in ~50 lines. The `build-w2a-sensor` skill walks an AI coding agent through discovery, signal design, scaffolding, and the install recipe — install it with:
  98. ```bash
  99. npx skills add https://github.com/machinepulse-ai/world2agent/skills/build-w2a-sensor
  100. ```
  101. Once it's ready, ship it to npm:
  102. ```bash
  103. npm publish
  104. ```
  105. That's it — your sensor is now installable by anyone, anywhere.
  106. ## Roadmap
  107. * **Graph layer** — compose and enrich signals from multiple sensors before they reach your agent. → [RFC](./docs/rfc-graph.md)
  108. ## Contributing
  109. * 🔧 **Build a sensor** — `npm publish` and it's live
  110. * 🐛 **Report bugs** — [open an issue](https://github.com/machinepulse-ai/world2agent/issues)
  111. * 💡 **Suggest a sensor** — [Discussions](https://github.com/machinepulse-ai/world2agent/discussions)
  112. → [Contributing guide](./docs/CONTRIBUTING.md)
  113. ## Community
  114. [Website](https://machinepulse.ai/) · [X / Twitter](https://x.com/MachinePulse_AI) · [YouTube](https://youtube.com/@MachinePulse_AI) · [Discord](https://discord.gg/hDjaD8pX)
  115. <!-- Star History — uncomment after launch -->
  116. <!-- [![Star History Chart](https://star-history.dera.page/svg?repos=machinepulse-ai/world2agent&type=Date)](https://star-history.dera.page/#machinepulse-ai/world2agent&Date) -->
  117. ## License
  118. [Apache 2.0](./LICENSE)
  119. ***
  120. <p align="center">
  121. Built by <a href="https://machinepulse.ai">MachinePulse</a> · Open source, open protocol, open invitation.
  122. </p>

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.