cv-generator

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill ingests untrusted data from external sources, presenting an indirect prompt injection surface.
  • Ingestion points: LinkedIn URLs, personal portfolio websites, and user-uploaded documents (CVs, Job Descriptions).
  • Boundary markers: No specific structural delimiters are defined to isolate external data content.
  • Capability inventory: The agent performs network fetches, document parsing, and OCR processing.
  • Sanitization: Employs a mandatory 'anti-hallucination enforcement gate' that blocks output if details cannot be verified directly from the source data.
  • [EXTERNAL_DOWNLOADS]: Retreives profile and portfolio information from user-specified URLs. This is an intended functionality for data aggregation.
  • [DATA_EXFILTRATION]: Processes sensitive personal information from resumes and profiles. The data flow is directed toward generating a user-facing document; no unauthorized transmission to third-party servers was detected.
  • [COMMAND_EXECUTION]: Instructions include performing OCR on scanned PDFs. This uses standard platform tools for document processing and does not involve arbitrary command execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 04:37 PM
Security Audit — agent-trust-hub — cv-generator