swain-dispatch

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

• Third-party content exposure detected (high risk: 0.90). The skill creates GitHub Issues and the required workflow explicitly triggers on and inspects issue bodies and issue comments (e.g., github.event.comment.body containing "@claude"), so untrusted public issue comments or issue content could be read by the Claude Code Action and influence agent behavior.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill triggers GitHub workflows that run the external GitHub Action anthropics/claude-code-action@v1 (https://github.com/anthropics/claude-code-action), which is fetched and executed at runtime and will process the dispatched issue body as agent instructions, making it a required external dependency that executes remote code and controls prompts.