hunter-resume
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses
python3to run a local scriptats.pyfor scoring resumes and validating content. This is a standard use of local processing for the skill's primary task. - [INDIRECT_PROMPT_INJECTION]: The skill processes job descriptions, which are external data sources. It mitigates potential prompt injection or hallucination risks by using a specialized 'truth check' tool (
ats.py trace) that compares the rewritten resume against the original 'source of truth' resume and user-confirmed facts. - [DATA_EXPOSURE]: The skill processes resumes and contact information but restricts operations to the local filesystem (
hunter/directory). The instructions explicitly forbid unauthorized data exfiltration or automated submissions.
Audit Metadata