job-hunt-fetcher

Pass

Audited by Gen Agent Trust Hub on May 4, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through the processing of untrusted job description screenshots.\n
  • Ingestion points: The skill ingests external data from user-provided images passed via the screenshots context variable (SKILL.md).\n
  • Boundary markers: There are no boundary markers, XML tags, or explicit instructions to treat extracted image text as untrusted content, which may lead the agent to follow instructions contained within the JDs.\n
  • Capability inventory: The skill has the capability to write files to the local filesystem at paths derived from work_dir. It constructs filenames using extracted data (company name and job title), creating a potential path traversal surface (SKILL.md).\n
  • Sanitization: The skill does not validate or sanitize extracted text before using it in file paths or markdown content, allowing malicious strings in screenshots to influence system operations.
Audit Metadata
Risk Level
SAFE
Analyzed
May 4, 2026, 04:10 PM
Security Audit — agent-trust-hub — job-hunt-fetcher