Tag: bedrock
All the articles with the tag "bedrock".
-
AgentCore Code Interpreter: When the Agent Has to Actually Run the Math
An agent will write correct-looking Python and then hand you a plausible number it never computed. AgentCore Code Interpreter is the managed sandbox that makes the agent run the code instead of guessing the output. I wired it into a real analyst task — computing the Pareto frontier of 500+ LLMs from a dated snapshot — and read the trace to see where the computation actually happens.
-
AgentCore Browser: A Managed, Isolated Browser an Agent Can Drive
Some tasks only exist on a screen. The reflex is to run Playwright on a box. AgentCore Browser is the managed alternative: a per-session, isolated Chromium you drive over CDP, with session isolation, live-view human takeover, and replay-to-S3 audit built in. Here's what the primitive is, how you reach it, and the ceiling it doesn't cross.
-
Beyond Semantic Retrieval: Metadata Filtering in Agent Memory
Semantic search finds things that sound related. Sometimes you need a WHERE clause instead — filter by priority, by date, by category. I tested metadata filtering in AgentCore Memory and found a quiet gate: custom keys silently drop unless declared as indexed keys. Here's what works and what decides whether any of it works at all.
-
Multi-Agent Memory: Agents That Actually Share Context
Sharing context between agents usually means pasting one agent's output into another's prompt. That doesn't scale and has no isolation boundary. AgentCore Memory offers a different mechanism: one shared memory store, actor-namespaced records, and semantic retrieval scoped to a namespace prefix. I ran it across two agents and three actors to see if it actually works.
-
Bedrock Managed Knowledge Bases as a Native Gateway Tool
AgentCore Gateway now ships a native bedrock-knowledge-bases connector that replaces the Lambda wrapper entirely. Attach the KB to the Gateway and it auto-exposes Retrieve and AgenticRetrieveStream over MCP. The constraint: it only works with Managed Knowledge Bases, not customer-managed ones.
-
What the Trace Knew That the Logs Didn't
Eight lessons from testing AgentCore observability on deployed agents, including what distinguishes broken agents from agents that print the right answer, why uninstrumented tool calls are invisible, and how to spot multi-agent failures that no single agent's logs can see.
-
AgentCore Registry Is a Governed Catalog for Agent Interfaces
Hardcoded runtime ARNs are the visible pain, and Registry does fix them — but agent discovery is fundamentally a governance problem, not a routing problem.
-
An Agent That Can Pay for Its Tools
Payment is not a tool call—it is a governed financial action. An agent that pays requires user consent, credential isolation, session budgets, policy checks, and a proof path the model cannot bypass.