hunt-file-upload
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow for interacting with untrusted data, which presents a surface for indirect prompt injection where malicious inputs from a target could influence agent behavior.\n- Ingestion points: Processing of file uploads, avatars, attachments, and structured archive files like ZIP and DOCX (SKILL.md).\n- Boundary markers: The instructions lack explicit delimitation or warnings to isolate processed external data from the agent's internal control logic.\n- Capability inventory: The skill utilizes capabilities for network requests (SSRF), file system manipulation (path traversal), and shell command verification (RCE) based on analyzed inputs.\n- Sanitization: No specific mechanisms for sanitizing or validating external content are included to prevent embedded instructions from being parsed as agent directives.
Audit Metadata