infrastructure-claude-code-game-workflow

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly configures MCP servers (see mcp-setup.md and templates/.mcp.json) — including a GitHub MCP for listing/reading issues and PRs and a Context7 MCP for pulling public library documentation — and the workflow examples require the agent to read and act on that external, user-generated/public web content, which could supply instructions that materially influence tool use and decisions.