Terms of Service

The Janitor Agent ("The Service") uses probabilistic Large Language Models (LLMs). You agree that you are the "Final Author" of all code generated.

• No Warranty on Code: We do not guarantee that code generated by the Fixer or Auditor is bug-free, secure, or performant.
• Auto-Merge Risk: By enabling "Auto-Merge," you voluntarily assume all risks associated with unreviewed code deployment, including data loss or production outages.

To the maximum extent permitted by law, the creators of Janitor Agent shall not be liable for any indirect, incidental, special, consequential, or punitive damages.

You agree to defend, indemnify, and hold us harmless from any claims, damages, or legal fees arising from:

• Your use of the Service to modify production environments.
• Any security vulnerabilities introduced into your codebase by the agent.
• Violation of any third-party rights or laws by code generated by the agent.

The Service relies on third-party APIs (e.g., OpenAI, Anthropic, GitHub). We are not responsible for:

• Service outages or API changes by these providers.
• "Hallucinations" or errors originating from the underlying LLM models.
• Data breaches that occur on third-party infrastructure.

The Service is currently in Alpha. You agree not to use Janitor Agent in high-risk environments (e.g., medical devices, nuclear facilities, financial ledgers) where software failure could lead to physical injury or financial ruin.

We reserve the right to revoke API keys immediately without notice if we detect reverse engineering, abuse of the "Agent Council" logic, or competitive analysis.