openclaw
Comprehensive guide for installing, configuring, operating, and troubleshooting OpenClaw — a self-hosted, multi-channel AI agent gateway. Use when the user asks about OpenClaw setup, configuration, channel management (WhatsApp/Telegram/Discord/Slack/iMessage/etc.), model provider setup, Gateway operations, multi-agent routing, security hardening, troubleshooting, or any maintenance task related to their local OpenClaw installation. Also use when encountering errors from `openclaw` CLI commands or the Gateway daemon.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/bengii/bengii-gemini-fixWhat This Skill Does
The OpenClaw skill serves as the foundational management interface for your self-hosted AI agent gateway. It enables you to bridge disparate messaging platforms—such as WhatsApp, Telegram, Discord, and iMessage—into a unified, local infrastructure. This skill provides the command-line control and configuration management necessary to deploy, maintain, and troubleshoot the OpenClaw daemon. It acts as the primary orchestrator for your multi-agent architecture, managing everything from model provider API keys and environment variables to the secure execution of local tools. By leveraging the OpenClaw skill, users can ensure their local agents remain responsive, secure, and compliant with custom security policies.
Installation
To integrate the OpenClaw core management skill into your existing environment, execute the following command in your terminal. Ensure your ClawHub CLI is up to date before proceeding:
clawhub install openclaw/skills/skills/bengii/bengii-gemini-fix
Once installed, you can verify the installation by running clawhub list to confirm the skill is active. You may need to restart the Gateway daemon to apply configuration changes after initial deployment.
Use Cases
- Multi-Channel Routing: Configure your AI agents to listen on Telegram and respond on Slack simultaneously without redundant model calls.
- Agent Lifecycle Management: Use internal hooks to rotate session keys, manage DM isolation, and perform database compaction for long-running agent threads.
- Security Hardening: Implement strict sandboxing for untrusted code execution using the
exectool with configured allowlists and safe-binaries. - Troubleshooting: Identify connectivity issues between your local node and external message platforms using the built-in channel diagnostic tools.
- Automated Workflows: Set up cron jobs or Gmail Pub/Sub triggers to automate daily briefings or proactive agent notifications.
Example Prompts
- "OpenClaw, verify my Telegram channel status and show me any recent authentication errors in the logs."
- "Set up a new secure sandbox environment for my python exec tool and restrict access to the /var/sensitive directory."
- "List all active agent sessions across Discord and Slack, and show me which ones have pending approvals for the Lobster workflow."
Tips & Limitations
- Performance: Always run OpenClaw in a Docker-isolated environment if you are testing third-party plugins to prevent accidental system impacts.
- Token Management: Use the
secrets.mdreference to ensure your provider tokens are stored in the secure vault rather than plain-text configs. - Limits: The Gateway performance is dependent on your local hardware; for multi-agent routing with heavy LLM tasks, ensure at least 16GB of RAM is allocated to the node process.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-bengii-bengii-gemini-fix": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: network-access, file-write, file-read, external-api, code-execution