resume-screener
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process resumes, which are external, untrusted documents. An attacker could potentially embed instructions within a resume (e.g., using hidden text or specific formatting) to influence the agent's evaluation or ranking.
- Ingestion points: Resumes provided by the user are evaluated in Step 3 of the SKILL.md file.
- Boundary markers: The instructions provide structure for evaluation but do not explicitly instruct the agent to ignore or delimit potentially malicious instructions embedded within the resume content.
- Capability inventory: The skill performs natural language reasoning and categorization; it does not request network access, file writes, or system-level commands within the instructions.
- Sanitization: No specific sanitization or filtering of resume content is performed by the skill.
Audit Metadata