test
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION] (SAFE): The skill utilizes local shell commands such as
tldr diagnosticsandtldr change-impact. These are used for static code analysis and environment verification within the project directory. - [REMOTE_CODE_EXECUTION] (SAFE): There are no indicators of remote script fetching (e.g., curl/wget to external domains) or execution of untrusted external code.
- [DATA_EXFILTRATION] (SAFE): Analysis of the workflow shows no access to sensitive system files (credentials, SSH keys) or unauthorized network transmissions.
- [PROMPT_INJECTION] (SAFE): The agent instructions are procedural and focused on test orchestration, containing no attempts to override safety filters or extract system prompts.
Audit Metadata