ai-social-media-content

Fail

Audited by Gen Agent Trust Hub on Mar 8, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill provides installation instructions that fetch and execute a shell script from the vendor's domain using curl -fsSL https://cli.inference.sh | sh.
  • [COMMAND_EXECUTION]: The skill is configured to use the Bash tool restricted to infsh commands. It provides numerous templates for running AI models through this CLI tool for video, image, and text generation.
  • [PROMPT_INJECTION]: The skill's workflow examples for batch processing and content repurposing exhibit an indirect injection surface by interpolating variables directly into shell commands.
  • Ingestion points: User-defined shell variables such as CONCEPT and TOPICS in the SKILL.md workflow examples.
  • Boundary markers: None; the variables are placed directly inside double-quoted strings in bash commands.
  • Capability inventory: The skill uses infsh app run which can access various external AI services.
  • Sanitization: No explicit sanitization or input validation is demonstrated in the provided shell templates.
Recommendations
  • HIGH: Downloads and executes remote code from: https://cli.inference.sh - DO NOT USE without thorough review
Audit Metadata
Risk Level
HIGH
Analyzed
Mar 8, 2026, 02:51 AM