Terms & Disclaimer

Last updated: August 3, 2026

1. What AgentWake Is

AgentWake is a free, open-source project by Benjamin Banwart ("we," "us," "our"). It is a single markdown file that instructs Claude Code to guide you ("you," "your") through setting up an autonomous AI agent on your macOS machine. Claude Code automates what it can and walks you step by step through the parts that need your input, such as macOS settings and creating a Discord bot. It is distributed free of charge under the MIT License.

2. License

AgentWake is licensed under the MIT License. You are free to use, copy, modify, and redistribute it under those terms. The full license text is in the project repository. Any brain files, configurations, and outputs your agent generates on your machine are yours.

3. Assumption of Risk

By using AgentWake, you acknowledge and accept the following:

• It configures an AI agent with full system access on your machine — the ability to read, write, and delete files, execute shell commands, install software, make network requests, and control a web browser — all without per-action approval prompts.
• The agent operates autonomously and may take actions that are unintended, incorrect, or harmful to your system, files, or data.
• The agent responds to Discord messages, meaning anyone with paired access can instruct it to act on your machine.
• AI systems make mistakes, including misinterpreting instructions and executing commands with unintended consequences.

We strongly recommend:

• Running AgentWake on a dedicated machine (e.g., a Mac Mini or VM), not your primary workstation.
• Keeping regular backups of important data.
• Restricting Discord access to trusted users only.
• Monitoring the agent's activity, and not storing sensitive credentials or personal data on machines running autonomous agents.

4. No Warranty

AgentWake is provided "as is," without warranty of any kind, express or implied, including but not limited to the warranties of merchantability, fitness for a particular purpose, and non-infringement. We do not warrant that it will be error-free, uninterrupted, or that the agent will behave as expected in all circumstances.

5. Limitation of Liability

To the maximum extent permitted by law, in no event shall the author be liable for any claim, damages, or other liability — including data loss, system damage, unintended actions taken by the agent, security breaches, or any direct, indirect, incidental, or consequential damages — arising from or in connection with AgentWake or its use.

6. Your Responsibility

You are solely responsible for the security of your machine and Discord bot; any actions the agent takes on your system or on third-party services; ensuring your use complies with all applicable laws and third-party terms; backing up your data; controlling who can communicate with your agent; and maintaining your Anthropic subscription in good standing.

7. Third-Party Services

AgentWake relies on third-party software and services, including Anthropic's Claude Code, Discord, Homebrew, and tmux. We are not affiliated with these services and are not responsible for changes to their availability, pricing, terms, or functionality. You are responsible for ensuring your use complies with Anthropic's Acceptable Use Policy and Terms of Service, which may change at any time. We make no guarantee that autonomous-agent use cases will remain permitted under those terms.

8. Governing Law

These terms are governed by the laws of the State of Kansas, without regard to its conflict-of-law provisions. If any provision is found unenforceable, the remaining provisions remain in full force and effect.

9. Changes

We may update these terms from time to time and will update the "Last updated" date above when we do.

10. Contact

Questions? Contact benjibanwart@proton.me.