cmux-debug
Pass
Audited by Gen Agent Trust Hub on Jul 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions to execute shell commands within cmux terminal panes using the
cmux sendcommand. This is used to trigger CLI actions and capture their output for debugging purposes. - [PROMPT_INJECTION]: The skill ingests untrusted data from terminal output through the
cmux read-screenandcmux capture-panecommands. This data is then processed by the agent which also has command execution capabilities viacmux send. - Ingestion points: Captured terminal output (SKILL.md)
- Boundary markers: None identified
- Capability inventory: Command execution via
cmux send(SKILL.md) - Sanitization: No sanitization of terminal output is specified before processing.
- [SAFE]: The skill's operations are confined to interacting with the cmux terminal multiplexer. It does not perform unauthorized network communications, access sensitive credentials, or employ obfuscation techniques.
Audit Metadata