Full Role Analysis Orchestrator
Pass
Audited by Gen Agent Trust Hub on Aug 3, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface due to the way it processes external and user-provided data.\n- Ingestion points: Untrusted data enters the agent context through the 'Current JD' and 'Sample resumes' fields in Phase 1, and through 'WebSearch' results in Agents 4 and 5.\n- Boundary markers: The subagent prompts (Phase 2) interpolate data like '[JD Text]' and '[RESUMES]' directly into instructions without using XML tags, triple quotes, or other delimiters to separate instructions from data.\n- Capability inventory: The skill uses parallel subagents capable of performing web searches and writing results to local markdown files (e.g., RECRUIT-ANALYSIS-[Role].md).\n- Sanitization: The instructions do not specify any sanitization, filtering, or validation for the content retrieved from resumes or the web.
Audit Metadata