agent-skill-management

Pass

Audited by Gen Agent Trust Hub on Aug 21, 2026

Risk Level: SAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill facilitates the use of the skills management tool via npx. This involves downloading and executing a utility provided by Vercel Labs, which is a well-known service provider. The instructions specify automated installation and refresh cycles to maintain the agent's capability suite.
  • [COMMAND_EXECUTION]: The package includes a Node.js maintenance script, scripts/check-installed-copies.mjs. This script is intended to be executed locally to perform integrity checks by comparing the contents of the skill source directory with the installed skill root. It uses standard, built-in Node.js modules for file system operations and does not perform any network requests.
  • [SAFE]: The skill promotes secure development practices by establishing a source-of-truth model for agent instructions. The inclusion of a drift detection mechanism provides a layer of protection against untracked changes in the agent's execution environment, ensuring that the instructions follow the project's version control.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 21, 2026, 10:17 PM
Security Audit — agent-trust-hub — agent-skill-management