NEWS · COMPANIES · #3382
Amazon Bedrock AgentCore gains Web Search API; LangChain/LangGraph integration
AWS announced a Web Search API for Amazon Bedrock AgentCore that returns signed, citeable web results from Amazon’s own index and supports domain allow/block lists and date-range filtering. LangChain’s AgentCore web search tool now calls this API and the post demonstrates a LangGraph daily dependency-watcher example using Anthropic Claude in Bedrock and AgentCore Gateway for multi-team deployment.
KEY POINTS
- AWS announced a Web Search API for Amazon Bedrock AgentCore that returns signed, citeable web results from Amazon’s own index and supports domain allow/block lists and date-range filtering.
- LangChain’s AgentCore web search tool now calls this API and the post demonstrates a LangGraph daily dependency-watcher example using Anthropic Claude in Bedrock and AgentCore Gateway for multi-team deployment.
- Provides agents with a built-in, signed, citeable web search and admin controls, enabling up-to-date, auditable agent workflows without routing queries to third-party search engines.
DECISION BRIEF
CONFIDENCE
WHAT CHANGED
Amazon Bedrock AgentCore added a Web Search API that returns signed, citeable web results from an Amazon-run web index and supports domain allow/block lists and publication date-range filtering. LangChain’s AgentCore web search tool now calls that API; the blog demonstrates a LangGraph daily dependency-watcher example run on AgentCore runtime and distributed via AgentCore Gateway.
WHY NOW
Agents need current web information (e.g., advisories, releases, prices) that model training data cannot provide. The new API supplies current, citeable search results inside AWS (signed with AWS credentials) and enforces source/date controls and admin policies, enabling auditable, production-ready agent workflows without routing queries to third-party search engines.
WHO IS AFFECTED
AI/agent developers and teams running agents on AWS (AgentCore runtime, Lambda, or ECS), LangChain developers and LangGraph users, operators using AgentCore Gateway and administrators who set source policies, and security/dependency monitoring workflows illustrated by the daily dependency-watcher example.
CONFIRMED
• AWS announced a Web Search API in Amazon Bedrock AgentCore that returns current web results signed with AWS credentials and served from an Amazon-operated web index. • Each search result includes the page title, URL, publication date, and an extract relevant to the query. • Requests can include or exclude domains and limit results to a publication date range. • Agents running on AWS Lambda, Amazon ECS, or Amazon Bedrock AgentCore call the API with their execution role (no separate search API key to store). • Web Search is available through AgentCore Gateway, where an administrator source policy can cover agents and MCP clients. • LangChain’s AgentCore web search tool now calls the Web Search API and can enforce rules (date window, source blocklist, search budget) that the model cannot change; it returns results as data for programmatic checks. • The AWS blog shows a LangGraph daily dependency-watcher example that ran three parallel branches in 36 seconds versus 99 seconds when run sequentially.
UNCERTAIN
• The excerpts do not provide API pricing, rate limits, or long-term data retention and logging policies for the Web Search API. • No third-party verification of the Amazon-run web index, indexing coverage, or ranking behavior is provided in the excerpts. • Security, privacy, and compliance details (e.g., audit logs retention, access controls beyond 'signed with AWS credentials', data residency) are not specified in the excerpts. • Broader performance at scale and multi-tenant behavior beyond the single LangGraph test (36s vs 99s) are not documented in the provided text. • The excerpts do not report adoption metrics, community or LangChain code changes beyond ’the tool now calls the API’, or timelines for broader ecosystem integrations. (Locations of missing evidence: API pricing/limits, index coverage, audit/logging/compliance specifics, large-scale benchmarks, adoption metrics.)
WHAT TO WATCH
Concrete observable signals to watch next: • AWS documentation or pricing pages publishing Web Search API pricing, rate limits, and retention/audit details. • Additional AWS blog posts or release notes with benchmark results or scale/performance metrics for Web Search and AgentCore Gateway. • LangChain and LangGraph repositories or release notes showing code changes, version updates, or examples using the AgentCore web search tool. • AgentCore Gateway admin policy documentation or examples of multi-team deployments beyond the provided example. • Third-party analyses or independent tests describing index coverage, relevance, and privacy/security behavior of the Amazon-run web index.
WHY IT MATTERS
Provides agents with a built-in, signed, citeable web search and admin controls, enabling up-to-date, auditable agent workflows without routing queries to third-party search engines.
EVIDENCE MAP
4Editorial claims linked to specific sources, with support, contradiction and context shown separately.
AWS announced a Web Search API for Amazon Bedrock AgentCore that returns signed, citeable web results from an Amazon-operated web index and does not send queries to a third-party search engine.
SUPPORTEDChecked 2026-11-11 · 1 supporting
Each search result returned by the API includes the page title, URL, publication date, and an extract relevant to the query; requests can include or exclude domains and limit results by publication date range.
SUPPORTEDChecked 2026-11-11 · 1 supporting
Agents call the Web Search API using their AWS execution role; there is no separate search API key to store or rotate.
SUPPORTEDChecked 2026-11-11 · 1 supporting
LangChain's AgentCore web search tool now calls the Bedrock AgentCore Web Search API, returning cited results and enforcing rules such as a date window, a source blocklist, and a search budget; it works with LangChain agents, LangGraph, and Deep Agents.
SUPPORTEDChecked 2026-11-11 · 1 supporting
SOURCES & TIMELINE
1AI agents increasingly handle work that depends on what changed today, such as security advisories, product releases, prices, and regulations. A model’s training data can’t answer those questions, so agents need web search. For production use, teams also need results they can cite, and control over which sources count. They want search queries to stay inside their cloud environment because those queries often reveal …
At AWS, we believe AI builders should be able to choose the right model for every task, trust their agents in production, and see clearly what their AI workloads cost. That belief guides how we invest across our portfolio. Amazon Bedrock gives you access to leading models and the tools to build and scale AI applications securely. We extend this foundation with Amazon Bedrock AgentCore , which provides the platform to…
AI agents and applications that answer questions about recent events, product releases, prices, or regulations need information newer than their model’s training data. Until now, Web Search on Amazon Bedrock AgentCore delivered that information to agents through AgentCore Gateway , a capability of Amazon Bedrock AgentCore. That means creating a gateway and a web search target and calling the tool from an MCP client. …