poster

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes shell commands through authorized tools to manage output, specifically using mkdir to create structured directories for assets and date to generate unique timestamps for file naming.
  • [EXTERNAL_DOWNLOADS]: The skill uses curl to download generated poster images from remote URLs provided by the image generation tool to the local file system.
  • [EXTERNAL_DOWNLOADS]: The documentation provides instructions for users to install the @blockrun/mcp package via npx. This is identified as a vendor-provided resource necessary for the skill's operation.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because user-supplied strings, including title, tagline, and subject, are interpolated directly into the master prompt template used to instruct the image generation model. This could be used to attempt to bypass the image model's safety filters.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 09:22 PM
Security Audit — agent-trust-hub — poster