Skills
▥AGENT LAB: SKILLS
skills/inference-sh-4/skills/ai-music-generation/Gen Agent Trust Hub

ai-music-generation

Fail

Audited by Gen Agent Trust Hub on Feb 19, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • Remote Code Execution (HIGH): The skill documentation includes the command curl -fsSL https://cli.inference.sh | sh. This pattern downloads a shell script from a remote URL and executes it immediately without verification. The domain inference.sh is not on the list of trusted external sources.
  • External Downloads (MEDIUM): The skill relies on an external CLI tool (infsh) that must be downloaded from a non-whitelisted domain to function.
  • Indirect Prompt Injection (LOW): The skill takes user-provided natural language text (prompts and lyrics) and interpolates them into JSON payloads for the infsh command-line tool.
  • Ingestion points: The --input JSON parameter in infsh app run commands (e.g., SKILL.md).
  • Boundary markers: Absent. The prompt is wrapped in quotes but not sanitized for escaping sequences.
  • Capability inventory: The skill uses the Bash tool to execute external CLI commands, which can interact with the network and local file system.
  • Sanitization: None detected. The agent is expected to insert user input directly into the command string.
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 19, 2026, 12:54 AM