cmux-debugging
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides architectural guidelines and debugging instructions for the cmux application. All referenced resources, including GitHub issue links, belong to the official vendor repository.
- [COMMAND_EXECUTION]: The skill contains a shell command for monitoring debug logs:
tail -f "$(cat /tmp/cmux-last-debug-log-path 2>/dev/null || echo /tmp/cmux-debug.log)". This command is intended for local development use and interacts with temporary files generated by the application's build scripts. - [INDIRECT_PROMPT_INJECTION]: The skill identifies a surface for processing developer queries and log data to assist in debugging. This is a standard functional requirement for a developer-oriented agent skill and is performed within the local environment scope.
- Ingestion points: Local log files accessed via tail and developer queries (SKILL.md).
- Boundary markers: Not specified.
- Capability inventory: Shell command execution via tail (SKILL.md).
- Sanitization: Not implemented.
Audit Metadata