GitHub for Beginners

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

• Third-party content exposure detected (high risk: 0.85). Outsider free text from GitHub issue/PR comments (e.g., the user mentions @claude ...) is ingested by the Claude Code workflow at runtime via the GitHub Actions/Claude Code integration, providing attacker-controlled prompt text.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly relies on the Claude GitHub App (https://github.com/apps/claude) and GitHub Actions/Anthropic API to run Claude Code at runtime (via @claude mentions), which causes an external service to execute code and make commits in the repository, so this URL represents a runtime external dependency that executes remote code.