Legal
Agent Risk Disclaimer
ClawPilot helps deploy OpenClaw-based agents. Agent systems can be powerful and risky. Read this before enabling production automations.
1. AI Outputs Can Be Wrong
Agent outputs can be inaccurate, incomplete, outdated, or misleading. You must review critical outputs before acting on them.
2. Tool Access Increases Risk
When agents have access to external tools, channels, and APIs, they can trigger real actions. Misconfiguration or prompt misuse may lead to unintended behavior, including unwanted outbound messages or data operations.
3. You Must Configure Safeguards
Use approval gates, allowlists, access boundaries, and monitoring before running live automations. Do not grant broad permissions that are unnecessary for your workflow.
4. High-Risk Use Requires Human Oversight
Do not use autonomous agents as the sole control for legal, medical, financial, safety, hiring, identity, or irreversible business decisions.
5. Third-Party Dependencies
ClawPilot depends on third-party model providers, communication channels, and APIs. Outages, policy changes, and rate limits in those systems can affect your agents.
6. OpenClaw Is Separate Software
OpenClaw documentation and license terms apply to the underlying software. You are responsible for complying with all relevant licenses, laws, and third-party terms.
7. No Guarantee
We do not guarantee uninterrupted availability, error-free outputs, or fitness for your specific use case. Use ClawPilot with appropriate operational controls.