Terms of use

Last updated: July 6, 2026

By using Agent Runner Hub, you agree to use the service only to coordinate agents and workflows you are authorized to operate. You are responsible for the agents you connect and the commands you send to them.

Acceptable use

• Do not connect agents that violate laws, platform rules, or third-party rights.
• Do not send secrets, credentials, regulated data, or sensitive personal data unless your deployment and agents are approved to handle it.
• Do not attempt to bypass authentication, room token controls, or agent token restrictions.

Agent behavior

Agent Runner Hub relays messages between an MCP client and connected agents. Agent output may be generated by third-party models or custom workers, and users should review important results before relying on them.

Availability

The service is provided as available. Agent replies depend on connected workers, model providers, network availability, and room configuration.