Amazon launches Web Answer on Bedrock AgentCore, priced at $10 per 1,000 calls
Amazon announced the general availability of Web Answer, a managed connector for Bedrock AgentCore that returns a synthesized, source‑cited answer to a natural‑language question in a single tool call. The service runs its own web searches, verifies facts, and formats citations, eliminating the need for developers to build a separate search‑retrieve‑synthesize pipeline.
Key points
- Web Answer returns a cited answer in one request, handling search, synthesis, and attribution
- Available in US East (N. Virginia) with a 10‑transactions‑per‑second per‑account limit
- Pricing is $10 per 1,000 tool calls on Amazon Bedrock AgentCore
Web Answer is accessed through an AgentCore Gateway, which authenticates calls via AWS IAM or JWT and invokes the tool using a service‑role permission (bedrock-agentcore:InvokeWebAnswer). It is currently available in US East (N. Virginia) with a rate limit of 10 transactions per second per account, and it costs $10 per 1,000 tool calls. The feature supports optional domain filters, date ranges, and localization, and returns results as markdown with inline citation markers.
The addition simplifies building agents for use cases such as regulatory Q&A, clinical support, customer service, and legal research, where up‑to‑date, auditable answers are required. Developers can add the tool to an existing gateway in minutes, then call it from any MCP‑compatible client such as Strands, LangChain, or CrewAI.
Getting started with Web Answer on Amazon Bedrock AgentCore
AWS Machine Learning Blog · 5 November 2026
Loading the full article…
This text was published by AWS Machine Learning Blog and written by Adrianna Chudy. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Agents & Tools
All →- Microsoft Foundry Local runs Qwen3.8-27B locally with GitHub Copilot on Nvidia GPUs · 1 src
- OpenAI 'dots' verified for planning automation; SBI Securities to offer Anthropic IPO applications · 1 src
- Meta's Muse composes song and produces vertical lyric video · 1 src
- OpenAI, Meta and xAI launch AI agents for long-term task delegation · 5 src
- Meta launches AI fashion app Style Sketch in Indonesia, Singapore and Thailand · 1 src
Comments
via GitHub Discussions