renoise-setup

Fail

Audited by Gen Agent Trust Hub on Sep 30, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDYNAMIC_EXECUTIONPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches manifest files, checksums, and binary archives from external sources, including the vendor's domain (download.renoise.ai) and a public GitHub repository (ArcoCodes/renoise-plugins-official).
  • [REMOTE_CODE_EXECUTION]: The skill executes native binaries downloaded from the internet. The scripts/install-cli.mjs script extracts these binaries and runs them using execFileSync to verify their functionality. The SKILL.md instructions specify that the --ensure flag allows for these installations to be performed without explicit user prompts in certain automated workflows.
  • [COMMAND_EXECUTION]: The skill executes various system commands and provides instructions for the user to run others. It uses execFileSync to run tar, node, powershell.exe, and the downloaded renoise binary. It also suggests installation commands using brew, apt, and winget for system dependencies.
  • [DYNAMIC_EXECUTION]: The installation process involves dynamic platform detection and runtime execution of downloaded code. The script install-cli.mjs determines the host OS and architecture at runtime to select the appropriate binary for download and execution.
  • [PRIVILEGE_ESCALATION]: The skill includes instructions that involve elevated permissions. It suggests using sudo apt install for dependencies on Debian-based systems. The installation script also uses chmodSync to modify file permissions, ensuring the downloaded binary is executable.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes structured data from an external source which influences its execution flow. The script scripts/install-cli.mjs ingests a JSON manifest from a remote server. While the script includes validation for schema versions and uses regex for asset names, the data retrieved determines which binary is downloaded and executed.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Sep 30, 2026, 12:51 PM