reel-cover-generator

Pass

Audited by Gen Agent Trust Hub on May 12, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [SAFE]: The skill requests absolute file paths for local image assets and output destinations. This is a functional requirement for the Gemini MCP tool used by the skill to access and save files on the host system. Users are guided on how to provide these paths safely.
  • [PROMPT_INJECTION]: The skill ingests untrusted text from user scripts to generate titles and subtitles for an image generation prompt. This creates a surface for indirect prompt injection. However, the implementation includes a mandatory review step (Step 5) where the full prompt is shown to the user for approval before any generation tools are invoked, providing effective mitigation.
  • [EXTERNAL_DOWNLOADS]: The documentation directs users to install external dependencies, specifically the @houtini/gemini-mcp package and the skill itself via npx. These are standard installation procedures for the skill's intended architecture and use of the Model Context Protocol (MCP).
Audit Metadata
Risk Level
SAFE
Analyzed
May 12, 2026, 09:31 PM
Security Audit — agent-trust-hub — reel-cover-generator