social-assets-factory

Pass

Audited by Gen Agent Trust Hub on Aug 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes several shell commands for project introspection and asset management. These include grep and find to discover branding styles, and sips (or ImageMagick) to normalize image dimensions. It also executes the codex CLI to perform image generation.
  • [EXTERNAL_DOWNLOADS]: The skill requires the OpenAI Codex CLI (@openai/codex), which is fetched from the official NPM registry or Homebrew. This dependency originates from a well-known organization.
  • [PROMPT_INJECTION]: The skill implements an 'Indirect Prompt Injection' surface by reading and interpolating data from local project files into image generation prompts.
  • Ingestion points: Local files such as DESIGN.md, BRAND.md, Tailwind configurations (tailwind.config.*), and various CSS or SCSS stylesheets found within the repository.
  • Boundary markers: None identified. The skill extracts branding tokens (colors, fonts, tone) and interpolates them directly into the prompt templates without using delimiters or instructions to ignore embedded commands.
  • Capability inventory: The skill possesses the ability to execute shell commands, manage local files, and invoke external image generation tools.
  • Sanitization: The skill does not programmatically sanitize the content extracted from project files. However, it instructs the agent to present extracted tokens to the user for confirmation or adjustment before generation, which serves as a manual verification checkpoint.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 12, 2026, 08:48 PM
Security Audit — agent-trust-hub — social-assets-factory