legal-data-hunter-pl

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

• Third-party content exposure detected (high risk: 0.75). SKILL.md describes a runtime workflow that fetches and normalizes legal data from multiple official/public portals/APIs (e.g., SAOS, Sejm ELI, courts), and that fetched document text (free-form content) would be ingested into the agent’s LLM context via the downstream “retrieve/normalize” pipeline.

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

• Potentially malicious external URL detected (high risk: 0.90). The SKILL.md explicitly instructs cloning and running code from the remote repository https://github.com/worldwidelaw/legal-sources.git (git clone … then running runner.py), so remote code would be fetched and executed as a required dependency.