skill-install-manager

Pass

Audited by Gen Agent Trust Hub on Aug 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill makes extensive use of shell commands to manage agent skills. It uses npx skills add, npx skills update, and npx skills ls to interact with the global skill environment. It also uses git ls-remote to check for remote version updates.
  • [EXTERNAL_DOWNLOADS]: The skill downloads content from various GitHub repositories listed in its reference files (e.g., github.com/anthropics/skills, github.com/vercel-labs/skills, github.com/MarecGents/marec-agent-skills). These targets include official repositories from well-known technology organizations and the skill's author, which are recognized as safe sources for these operations.
  • [REMOTE_CODE_EXECUTION]: While the skill downloads and installs other skills (which contain code), this behavior is the primary intended purpose of a 'Skill Manager'. It implements a fallback mechanism to manually download files from GitHub's raw content servers if the standard npx skills CLI fails. These operations are limited to the specific repositories defined in the skill's configuration and are not triggered by arbitrary user input.
  • [SAFE]: The skill follows security best practices by implementing strict timeouts (15-30 seconds) for all network and shell operations to prevent the agent from hanging. It also provides a clear reporting mechanism to the user detailing every action taken during the synchronization process.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 16, 2026, 04:02 AM
Security Audit — agent-trust-hub — skill-install-manager