Claway
Claway is Blockway's AI agent for teams. It chats with your staff and customers in WhatsApp, Telegram, Slack, Microsoft Teams and the Claway App, remembers context per chat, answers from your own documents, runs scheduled jobs, and connects to n8n, Zapier or your own software. You run it on a Claway Box in your office or as hosted Claway, and manage it from the OPS dashboard.
Start here
| Page | What it covers |
|---|---|
| Getting started | Claway Box or hosted, first sign-in to OPS, connectors, the Claway App |
| Operator guide | Running Claway for a team: personas, knowledge base, WhatsApp groups |
| End-user cheat sheet | One page for the people who chat with the agent every day |
| Claway App | The employee app for iOS, Android, desktop and web |
| Webhooks | Send a message from n8n, Zapier, Make or your code, get the answer back |
| API server | OpenAI-compatible /v1 API for n8n AI nodes, Open WebUI and SDKs |
| Messaging platforms | WhatsApp, Telegram, Slack, Teams, Email, WeCom, Feishu and more |
| Features | Tools, skills, memory, scheduled jobs, voice, images and browsing |
| Skills catalog | Everything the agent can learn to do, one page per skill |
| FAQ & troubleshooting | Common questions and fixes |
What Claway does
- Lives in your chat apps. One agent across WhatsApp, Telegram, Slack, Microsoft Teams, Email, WeCom, Feishu, DingTalk and more, plus the Claway App.
- One persona per chat. Each group gets its own role, memory, skills and knowledge-base scope, kept apart from every other group.
- Learns and remembers. It keeps memory per chat, writes reusable skills from experience, and searches past conversations.
- Does the work. Web search, browsing, documents, spreadsheets, images, voice, scheduled jobs and delegated sub-tasks.
- Connects to your systems. Webhooks and an OpenAI-compatible API for n8n, Zapier, Make and your own code, plus MCP servers for other tools.
- Under your control. OPS records changes in its audit log; connectors stay off until you enable them with a secret.
For LLMs and coding agents
Machine-readable entry points to this documentation:
/llms.txt: a curated index of the main pages with short descriptions./llms-full.txt: every page concatenated into one Markdown file.