AI Agents

OpenClaw standing orders let agents run defined programs safely. Learn scope, triggers, approval gates, escalation rules, examples, and where to store them.

Learn how OpenClaw agent sandboxing works, what it protects, where it stops, and how to set safer defaults for tool use.

OpenClaw secrets management keeps API keys out of agent-readable files. Learn SecretRefs, audits, snapshots, secret stores, and egress proxy trade-offs.

OpenClaw model failover keeps AI agents running through auth, rate-limit, timeout, and provider overload failures without hiding bad config.

Run OpenClaw local models with LM Studio, Ollama, or a custom GPU server. Learn config choices, safety tradeoffs, and setup checks before switching today.

Set up OpenClaw WhatsApp with QR login, sender allowlists, group controls, and practical troubleshooting for a private, self-hosted AI agent you control.

Set up OpenClaw Signal with signal-cli, pairing, group controls, media limits, and troubleshooting for a private self-hosted AI agent on your own server.

Set up OpenClaw Slack with Socket Mode, HTTP Request URLs, token choices, allowlists, mention gates, threads, slash commands, file handling, and fixes.

Set up OpenClaw Discord with bot tokens, intents, OAuth scopes, guild allowlists, pairing, slash commands, threads, and troubleshooting.