talking-head-production

Pass

Audited by Gen Agent Trust Hub on Sep 21, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions for installing the belt CLI tool from a GitHub repository (inference-sh/skills) and suggests adding related skills via npx commands.
  • [COMMAND_EXECUTION]: The instructions utilize the belt CLI tool to execute remote AI application runs for image and video synthesis.
  • [INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it interpolates user-controlled text into prompts and voice scripts for remote AI models. * Ingestion points: Parameters like voice_script, prompt, and video_prompt in SKILL.md examples. * Boundary markers: No specific delimiters are used to separate user input from instructional content in the examples. * Capability inventory: The skill executes remote subprocesses via the belt tool. * Sanitization: No explicit sanitization of input strings is performed within the provided instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 21, 2026, 08:14 PM
Security Audit — agent-trust-hub — talking-head-production