How Gen Digital built Norton Revamp’s AI-curated News Engine with Amazon Bedrock AgentCore Web Search
Gen Digital’s Norton Revamp platform launched in March 2026 uses Amazon Bedrock AgentCore Web Search to power an AI‑curated News Engine that delivers real‑time, cited content for professional reputation management. By sending natural‑language queries to the fully managed AgentCore Web Search connector, the system receives snippets, URLs, titles, and publication dates, which are then used to generate personalized news feeds and smart content suggestions. This approach eliminates the need for custom crawlers or third‑party news APIs, offering zero infrastructure overhead, zero data egress, and built‑in compliance with the Model Context Protocol.
⚡ Key Takeaways
- AgentCore Web Search provides real‑time snippets, source URLs, titles, and publication dates for each query.
- The capability is delivered as a managed connector on the AgentCore Gateway within Amazon Bedrock.
- Using AgentCore removes infrastructure maintenance, avoids data egress, and eliminates vendor‑managed news API contracts.
- Integration is done by sending natural‑language queries from the agent to the Web Search connector, receiving structured results for downstream LLM grounding.
- The solution requires operating within AWS and leveraging the Model Context Protocol‑compliant AgentCore service.
Engineers can add up‑to‑date, source‑cited knowledge to LLM‑driven applications without building or maintaining their own web crawlers, reducing latency, cost, and compliance risk for production AI services.
✅ Practical Steps
- Configure your Bedrock AgentCore instance to include the Web Search managed connector on the AgentCore Gateway.
- From your agent code, issue natural‑language queries to the Web Search connector and parse the returned snippets, URLs, titles, and publication dates for grounding LLM responses.
Want the full story? Read the original article.
Read on AWS ML Blog ↗