video-ad-specs

Fail

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • Remote Code Execution (HIGH): The skill contains the command 'curl -fsSL https://cli.inference.sh | sh', which downloads a script from an untrusted domain and executes it directly. This pattern is a major security risk as the remote script could be modified to perform malicious actions on the host system.
  • Command Execution (MEDIUM): The skill requires 'Bash(infsh *)' permissions to run CLI tools. While necessary for its advertised function, this provides an execution path for potentially malicious or vulnerable external binaries.
  • External Downloads (MEDIUM): The skill suggests installing further extensions via 'npx skills add'. These external packages are not verified and represent a supply-chain risk if the remote repository is compromised.
  • Indirect Prompt Injection (LOW): The skill processes user-supplied data to generate prompts for external AI models through the 'infsh' tool. Evidence Chain: 1. Ingestion points: Arguments in the 'infsh app run' command blocks. 2. Boundary markers: Absent (missing delimiters to isolate untrusted user data). 3. Capability inventory: Shell execution and network access via CLI. 4. Sanitization: None detected.
Recommendations
  • HIGH: Downloads and executes remote code from: https://cli.inference.sh - DO NOT USE without thorough review
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 17, 2026, 06:49 PM