This is a small, personal project. There is no formal support window, but I will do my best to respond to security-related issues on the latest tagged release.
If you believe you have found a security issue related to the scripts, prompts, or documentation:
- Please open a private issue or contact me via GitHub (preferred), and
- Avoid including any real secrets, tokens, or proprietary code in the report.
Because Aku Loop uses autonomous LLM agents and the claude CLI with --dangerously-skip-permissions, you should:
- Run it only on non-production machines and repositories.
- Use disposable branches or throwaway repos while you experiment.
- Store API keys and tokens in environment variables or local config files that are gitignored.
- Avoid running the loop under accounts that have broad, unnecessary access to cloud resources.
Logs (./logs/) may contain model responses and file paths. Handle them as you would handle any other developer logs (do not share them publicly if they contain sensitive details).