update-claude

Pass

Audited by Gen Agent Trust Hub on Aug 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes various shell commands (ls, find, cat, readlink, mkdir, ln, mv) to audit and modify the repository's configuration directory. These operations are scoped to the project environment and include checks for command availability.
  • [EXTERNAL_DOWNLOADS]: The skill utilizes npx skills add to fetch and update specialized agent skills from a vendor-owned repository (Fandhe-AI/agent-cli-skills). This is a documented functionality for extending the agent's capabilities using the author's provided toolset.
  • [COMMAND_EXECUTION]: The skill uses the GitHub CLI (gh) to verify authentication status and query the repository's API for issue data, facilitating integration with GitHub workflows.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: The skill includes explicit security guidance for the AI agent, warning against the inclusion of API keys, tokens, or passwords in shell commands or configuration files.
  • [DYNAMIC_EXECUTION]: The skill manages the lifecycle of workflow scripts by creating and verifying relative symlinks, ensuring that local modifications do not break the connection to managed skill assets.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 19, 2026, 12:09 AM
Security Audit — agent-trust-hub — update-claude