yungroai-skills
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill automates the discovery and installation of business-specific skills from the author's repository (
YunGroAI/agent-skills) using theskillsutility from Vercel Labs, which is a recognized service for managing agent extensions. - [COMMAND_EXECUTION]: It uses shell commands to interact with the host agent's environment, such as adding MCP servers (
claude mcp add) and managing the local skill synchronization lifecycle vianpx skills. All such operations are transparently defined and subject to the platform's execution permissions. - [SAFE]: The authorization model for the growth engine tools relies on standard MCP OAuth browser-based flows, explicitly avoiding the need for hardcoded API keys, secrets, or tokens within the agent's configuration files.
- [SAFE]: The skill implements a local 24-hour throttling mechanism for updates (storing a timestamp in
~/.cache/yungroai-skills/last-check), which is a standard development practice to minimize unnecessary network traffic and ensure efficient operation.
Audit Metadata