claude-code-plugin

Pass

Audited by Gen Agent Trust Hub on Oct 8, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a comprehensive documentation resource for plugin developers. The instructions are informative and follow the platform's standard development practices.
  • [EXTERNAL_DOWNLOADS]: Mentions official documentation URLs from Anthropic and references the author's own repository (miles990/evolve-plugin) as a practical example. These references are appropriate for the skill's instructional context and target trusted or vendor-owned locations.
  • [COMMAND_EXECUTION]: Provides example shell scripts for tasks like version synchronization and security hooks. For instance, the protect-env.sh example explicitly demonstrates a security best practice by blocking access to sensitive file extensions (e.g., .env, .key, .secret) during tool execution. These scripts are provided as static examples for the user to implement rather than being executed by the skill itself.
  • [DATA_EXFILTRATION]: References the storage locations for installed plugins and known marketplaces (e.g., ~/.claude/plugins/installed_plugins.json). These references are intended to help users debug or manually manage their installation and do not represent unauthorized data access or exfiltration patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 8, 2026, 09:40 PM