configure-claude-plugins

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). SKILL.md instructs creating GitHub Actions (.github/workflows/claude.yml and claude-code-review.yml) that set plugin_marketplaces to the public GitHub repo https://github.com/laurigates/claude-plugins.git, causing the Claude Code action to fetch and run third‑party plugins from an untrusted public marketplace which can materially influence agent behavior.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The workflow templates set plugin_marketplaces to https://github.com/laurigates/claude-plugins.git, which the Claude Code action will fetch at runtime to load plugin definitions that directly affect agent prompts/behavior (and may include executable plugin code), so this is a required runtime external dependency.