nimbalyst-skills

Warn

Audited by Gen Agent Trust Hub on Jun 25, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to download and install packages from a custom registry (nimbalyst.com/skills) using the npx skills add command. This introduces external code into the execution environment that is not subject to standard security verification.
  • [REMOTE_CODE_EXECUTION]: By directing the agent to use npx, the skill triggers the download and immediate execution of scripts from a remote source. This pattern is used to run unverified code from an external vendor.
  • [COMMAND_EXECUTION]: The instructions require the agent to have shell access to check for existing skill installations in the user's home directory path ~/.design-agent-skills/ and to execute installation commands.
  • [PROMPT_INJECTION]: The documentation includes instructions for the agent to use shell execution bypasses (such as the ! prefix in specific environments like Claude Code), which encourages the agent to perform actions outside of its standard constrained interface.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 25, 2026, 06:33 PM
Security Audit — agent-trust-hub — nimbalyst-skills