github-copilot-starter

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

• Third-party content exposure detected (high risk: 0.90). The skill's required workflow (see the "MANDATORY FIRST STEP" and "Research Strategy with fetch tool" in SKILL.md) explicitly instructs the agent to fetch and adapt content from the public GitHub repository https://github.com/github/awesome-copilot (e.g., README.instructions.md, README.agents.md, README.skills.md and related tree links), which are open/public third‑party sources the agent must read and use to drive file creation and decisions—exposing it to untrusted, user-provided content that could inject instructions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly mandates using the fetch tool at runtime to retrieve and incorporate instructions/agents from the awesome-copilot GitHub repo (e.g., https://github.com/github/awesome-copilot/blob/main/docs/README.instructions.md), and that fetched content would directly control agent prompts/instructions and is required for the skill to operate.