screen-coder

Pass

Audited by Gen Agent Trust Hub on Apr 12, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by ingesting untrusted data while possessing high-privilege tool capabilities.
  • Ingestion points: The skill accepts a path to a screenshot image as a command argument which is then analyzed by the agent.
  • Boundary markers: There are no explicit delimiters or instructions telling the agent to disregard potential instructions embedded within the visual content of the screenshot.
  • Capability inventory: The skill allows the use of Write and Bash tools, which could be exploited if an attacker embeds malicious commands in a design image.
  • Sanitization: No validation or sanitization process is described for the content extracted from the external image file.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 12, 2026, 11:56 AM
Security Audit — agent-trust-hub — screen-coder