screenshot-analyzer

Pass

Audited by Gen Agent Trust Hub on Apr 1, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by processing external screenshot data.
  • Ingestion points: Accesses image files residing in "~/Downloads/Screenshots/" via the "Read" tool as defined in "SKILL.md".
  • Boundary markers: Absent; there are no explicit delimiters or instructions provided to the agent to disregard or isolate embedded text or instructions found within the visual content of the screenshots.
  • Capability inventory: The skill utilizes shell commands ("ls") and file-reading capabilities ("Read" tool) across its workflow.
  • Sanitization: Absent; no mechanism is defined to sanitize, filter, or validate the content of the screenshots before they are processed by the model.
  • [COMMAND_EXECUTION]: The skill uses local shell commands to facilitate file discovery and management.
  • Evidence: The skill instructs the agent to execute "ls -t ~/Downloads/Screenshots/*.png | head -5" to identify and sort recent files for analysis.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 1, 2026, 06:50 AM