create-research-brief
Warn
Audited by Snyk on Mar 31, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 1.00). The skill explicitly directs models to perform web_search and ingest external community sources (e.g., references/model-profiles.md and the GPT-5.2 Site Restriction Library instruct site-restricted searches and list domains like github.com, stackoverflow.com, reddit.com, producthunt.com, news sites, etc.) and requires grounding analysis/citations from those third‑party pages, so untrusted web content can be read and materially influence prompts, decisions, and consolidation.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata