prompt-library

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

• Third-party content exposure detected (high risk: 0.85). The runtime pull command fetches prompt content from a (configurable) public API endpoint (api_get → entry.get("content")), and that fetched free-text body is then printed/used for downstream LLM context, so an outsider-hosted prompt library can inject arbitrary text.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's runtime fetches prompt content from https://broomva.tech/api/prompts (e.g., GET https://broomva.tech/api/prompts/) and directly uses the returned "content" as agent prompts/instructions, so this external endpoint controls agent behavior at runtime.