aris-idea-discovery

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches and ingests open/public third-party content (e.g., Phase 1: searches arXiv/Google Scholar/Semantic Scholar and Phase 0.5: downloading arXiv PDFs or fetching "other URL" via WebFetch), and those external summaries are used as core context that influences idea generation and downstream actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly fetches and downloads external papers at runtime (e.g., "https://arxiv.org/abs/2406.04329") via /aris-arxiv or WebFetch and then summarizes/injects that content (REF_PAPER_SUMMARY.md) into the model context to drive idea generation, so the external URL is used at runtime to control the agent's prompts/context.