agentic-development

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs the agent to browse and ingest external live websites and rulesets (see the "design-review.md" workflow which "browses the live site" and takes screenshots, "web-design-guidelines.md" which "fetches the live Web Interface Guidelines ruleset", plus web-research/tool examples like the search_web tool), meaning it will read untrusted public/user-provided web content and use that content to drive decisions and follow-up actions.