I Rebuilt a Browser Fact-Checker on AWS, and AgentCore Web Search Was the Missing Piece

Thu, 25 Jun 2026 00:00:00 +0000

🎬 Also available as a blog walkthrough video if you’d rather watch than read.

TL;DR: I came across an open-source browser extension that fact-checks videos and live streams in real time, and I wanted to know what it would take to rebuild that idea as an AWS-native service. The answer turned out to be short, because of one recently launched piece. The pipeline is the same everywhere: transcribe, pull out the checkable claims, verify each against the web, show a verdict. The hard part has always been the verify step, which needs a model that can search the live web and cite what it found. Web Search on Amazon Bedrock AgentCore, generally available since 19 June 2026, is a managed, MCP-compatible tool that gives exactly that to any model, including Claude. With it, the build is Cognito for login, API Gateway and Lambda for the seam, Claude on Bedrock for extraction and verdict reasoning, and AgentCore Web Search for grounded, dated, cited evidence. The code is on GitHub.

Disclaimer: I’m a solutions architect who builds things to understand them. Treat this as a builder’s field report, not authoritative guidance. The service is a proof of concept; verdicts are model-generated and you should always confirm against the cited sources. If I’ve got something wrong, tell me.

What Got Me Started

I came across an open-source browser extension that fact-checks YouTube videos and live streams in near real time, entirely in the browser [3]. The idea is clean and the pipeline is easy to describe: transcribe what is being said, identify the verifiable claims inside the transcript, verify each claim against the web, and highlight the result inline as the video plays. It leans on Google Gemini’s built-in search grounding for the verification step, and it does the job well for a single user with a personal API key.

Give Every Agent Its Own Computer: AWS Lambda MicroVMs

Wed, 24 Jun 2026 00:00:00 +0000

🎬 Also available as a blog walkthrough video where I walk through the post and the demo.

TL;DR: AWS Lambda MicroVMs (launched 22 June 2026) gives every user, job, or AI agent its own isolated, stateful compute environment: a Firecracker microVM with VM-level isolation, near-instant launch and resume from a snapshot, and suspend/resume state for up to eight hours, all in one API call with no virtualization infrastructure to operate. Until now you had to pick two of isolation, speed, and state, and run the plumbing for the rest yourself. I built a small demo where Amazon Bedrock writes Python and a per-session microVM runs it, keeps its state across a suspend, and stays sealed off from a second microVM. The code is on GitHub. This is the clean answer to “where does my agent’s code actually run?”

Disclaimer: I’m a solutions architect who builds things to understand them. This is a builder’s field report on a feature I tested and a demo I wrote, not authoritative guidance. If I’ve got something wrong, tell me.

The Idea That Stuck

At a machine learning conference last year, the idea that every AI agent should get its own computer stuck with me. Its own place to run code, keep some state, and not step on anyone else’s work. It made plain intuitive sense. An agent that writes and runs code is a lot less scary when that code executes somewhere boxed off from everything that matters.

schristoph.online

I Rebuilt a Browser Fact-Checker on AWS, and AgentCore Web Search Was the Missing Piece

What Got Me Started

Give Every Agent Its Own Computer: AWS Lambda MicroVMs

The Idea That Stuck