resume-panel
Pass
Audited by Gen Agent Trust Hub on Nov 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection because it reads and processes externally provided resumes and job advertisements. These inputs are inserted into subagent prompts without sufficient isolation or sanitization techniques.
- Ingestion points: The skill reads a resume from a file path (PDF, DOCX, MD, TXT) or pasted text, as well as a target job description or link, as defined in the 'Inputs' section.
- Boundary markers: No explicit delimiters or instructions are provided to help the subagents distinguish between the instructions of the skill and potentially malicious instructions embedded within the resume content.
- Capability inventory: The skill possesses the ability to read and write files to the local file system (creating a 'resume-panel/' directory and individual markdown files) and orchestrate multiple subagents via the Task tool.
- Sanitization: There is no evidence of filtering or sanitizing the input text before it is interpolated into the prompts for the seven reviewers.
Audit Metadata