resume-screener
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data (resumes and job descriptions) which could contain adversarial instructions designed to manipulate the screening outcome.
- Ingestion points: SKILL.md instructs the agent to ingest and process user-provided job descriptions and candidate resumes.
- Boundary markers: The skill lacks explicit instructions to use delimiters or 'ignore instructions' warnings when handling external text.
- Capability inventory: The skill has the capability to write structured screening reports to /mnt/user-data/outputs/screening-report.md.
- Sanitization: There is no evidence of input validation, filtering, or escaping of the text within the resumes before the agent performs its evaluation.
Audit Metadata