discover-testing
Pass
Audited by Gen Agent Trust Hub on Apr 23, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill includes several Python scripts in the
scripts/directory that usesubprocess.callto invoke other local scripts (e.g.,convert_formats.pyandparse_formats.py). These executions are used for internal orchestration of file format conversions and do not involve shell injection vulnerabilities or execution of untrusted remote content. - [EXTERNAL_DOWNLOADS]: No evidence of external network operations, script downloads, or package installations was found. The skill's functionality is contained within its local environment.
- [PROMPT_INJECTION]: The instructions and prompts provided in
SKILL.mdandprompts/discover-testing.mdare focused on task routing and do not contain attempts to override agent behavior or bypass safety filters. - [DATA_EXFILTRATION]: While the skill reads and processes various file formats, there are no network-capable tools or functions present that would allow data to be sent to external domains.
Audit Metadata