github-release-management

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches and ingests user-generated GitHub content (commit messages, PRs, discussions) via commands like "gh pr list --json ..." and "gh api repos/:owner/:repo/compare ..." (see "Progressive Disclosure: Level 1" and the GitHub Actions "Initialize Release Swarm" / "Generate Release Changelog" steps) and then uses that data to generate changelogs, version suggestions, PR bodies and drive automated releases/deployments, so untrusted third-party content can materially influence agent actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The workflow and examples repeatedly run "npx claude-flow@alpha" (which fetches and executes package code from the npm registry, e.g. https://registry.npmjs.org), so the skill at runtime pulls remote code that directly implements/controls the agent orchestration and is a required dependency.