gemini

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly enables Google Search grounding (e.g., the --grounded flag described in SKILL.md/README and implemented in scripts/gemini.py via types.Tool(google_search=types.GoogleSearch())), and the Brainstorm/grounded workflows (references/brainstorm.md and SKILL.md) treat web-checked "Verified Context" as input to subsequent reasoning, so arbitrary public web content can be ingested and materially influence downstream tool use and decisions.