skills/mnhkahn/cyeam-cli/mo/Gen Agent Trust Hub

mo

Pass

Audited by Gen Agent Trust Hub on Jul 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches a CLI binary from the author's repository (github.com/mnhkahn/cyeam-cli).
  • [COMMAND_EXECUTION]: Uses chmod +x to grant executable permissions to the downloaded binary.
  • [COMMAND_EXECUTION]: Employs sudo mv to install the tool into a system directory (/usr/local/bin), which requires administrative access.
  • [DATA_EXFILTRATION]: The OCR command (cyeam mo ocr) uploads local images to the vendor's domain (cyeam.com) for analysis.
  • [PROMPT_INJECTION]: The skill processes untrusted image data through OCR, presenting a surface for indirect prompt injection.
  • Ingestion points: Image file processing via the ocr command in SKILL.md.
  • Boundary markers: None specified for the OCR output.
  • Capability inventory: Executes external binaries and performs network uploads.
  • Sanitization: No explicit sanitization or filtering of OCR results is documented.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 18, 2026, 03:58 PM
Security Audit — agent-trust-hub — mo