job-hunt-fetcher
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests text parsed from user-provided screenshots and writes it into markdown files, creating a potential surface for indirect injection if malicious instructions are present in the image text. 1. Ingestion points: User-provided screenshots processed in SKILL.md. 2. Boundary markers: The skill uses a structured YAML and Markdown template to organize extracted data, though the text fields themselves contain raw extracted strings. 3. Capability inventory: File writing to the .work/jd-pool/ directory. 4. Sanitization: No specific sanitization or validation is performed on the extracted text before writing it to files.
- [SAFE]: No malicious patterns, obfuscation, or unauthorized network operations were detected. The skill operates within its stated purpose of screenshot parsing and localized file management.
Audit Metadata