new-skill

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md explicitly instructs the agent to fetch and read third‑party README content via curl from raw.githubusercontent.com (curl -s https://raw.githubusercontent.com/owner/repo/main/README.md), meaning the agent ingests untrusted public web content that can influence its decisions and tool use.

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

• Potentially malicious external URL detected (high risk: 0.90). The SKILL explicitly instructs the agent to fetch and read runtime content via the curl command (https://raw.githubusercontent.com/owner/repo/main/README.md), so external raw.githubusercontent.com content would be loaded into the agent's context and directly control prompts/behavior.