Termux-AI includes features for autonomous terminal operation, memory handling, validated shell execution, and optional voice interaction.

This project is intended for:

• personal learning
• local development
• controlled automation
• authorized security research only

Do not use the project on systems, networks, or devices you do not own or do not have explicit permission to test.

Humans keep treating automation like a substitute for judgment. It is not.

Only the current maintained version of the project is supported.

Older snapshots may still run, but they are not actively maintained and should not be treated as safe by default.

If you discover a security issue, do not open a public issue.

Instead, report it privately through the repository Security tab if available, or contact the maintainer directly through GitHub.

Please include:

• a clear description of the issue
• steps to reproduce it
• likely impact
• any suggested fix

Examples include:

• unsafe shell command construction
• command validation bypasses
• unauthorized file writes
• leaks involving api.keys
• unwanted access to core/
• broken permission boundaries
• unsafe logging or memory handling

These paths deserve extra care:

• api.keys
• core/
• Termux-STT/
• memories.txt
• log.txt

Do not commit secrets, tokens, or personal data into the repository. Git history remembers everything like a vindictive archivist.

Termux-AI is a terminal assistant, not a liability shield. If a feature can affect a system, network, or account, use it responsibly and only where you have permission.