skills/1nfsh/skills/video-ad-specs/Gen Agent Trust Hub

video-ad-specs

Fail

Audited by Gen Agent Trust Hub on Feb 18, 2026

Risk Level: CRITICALREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • Remote Code Execution (CRITICAL): The skill explicitly instructs users to execute curl -fsSL https://cli.inference.sh | sh. This is a high-risk pattern that executes arbitrary code from a remote, untrusted server directly on the host system without verification.
  • External Downloads (HIGH): The skill uses npx skills add to fetch multiple external components from inference-sh/skills. These dependencies are not from trusted sources like official GitHub organizations or verified registries.
  • Command Execution (MEDIUM): The skill relies on the Bash(infsh *) tool to execute complex operations. While this is the intended functionality, it provides a significant attack surface when combined with untrusted remote code.
  • Indirect Prompt Injection (LOW): The skill is vulnerable to indirect prompt injection through its video generation prompts.
  • Ingestion points: User-controlled strings passed to the --input flag in infsh commands within SKILL.md.
  • Boundary markers: Absent; there are no delimiters or instructions to ignore embedded commands in the input data.
  • Capability inventory: The skill can execute bash commands and interact with remote APIs via the infsh CLI.
  • Sanitization: None detected; user input is interpolated directly into JSON payloads for command execution.
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
CRITICAL
Analyzed
Feb 18, 2026, 01:59 AM
Security Audit — agent-trust-hub — video-ad-specs