ai-coding-agent-setup

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

• Third-party content exposure detected (high risk: 0.90). The skill's MCP documentation (resources/mcp-codebase-tools.md and the .vscode/mcp.json examples) explicitly configures a GitHub MCP Server and tools like get_file_contents and search_code that let the agent fetch and interpret public repository contents and PR/issue data (untrusted, user-generated) as part of its workflow, which can materially influence subsequent tool use and decisions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly recommends installing and configuring the GitHub MCP server (https://github.com/github/github-mcp-server), which at runtime provides tool calls like get_file_contents and search_code that fetch remote repository content into the agent context (directly controlling prompts/instructions and enabling remote-executing tool workflows), and the guide relies on this MCP server as a required capability for code navigation.