AWS details granular access control for Bedrock knowledge tools
AWS Machine Learning Blog published a guide to securing Amazon Bedrock Managed Knowledge Base tools for multi-tenant AI agents. Its architecture uses Amazon Bedrock AgentCore Gateway and Model Context Protocol tools, with trusted code injecting verified user attributes and overriding model-generated search context. Additional AgentCore policies and resource-based controls provide independent safeguards.
AWS Machine Learning Blog says this design keeps authorization decisions out of the LLM, reducing the risk that prompt injection or model errors could expose internal or customer data across access boundaries.
More enterprises are building AI agents into their daily work, and a common pattern is an agent that searches internal documents to compose its answers. A single organization might soon run hundreds of agents, each connected to many data sources and tools. This raises a question. When an agent searches on behalf of a user, can you make sure that it uses only the information that user is allowed to see?
Consider AnyCompany, a fictional company that builds an internal assistant shared by all employees. An employee in department A asks, “Compare my department’s business plan with the notice to all departments, and list the initiatives that are behind schedule.” The agent writes its own search queries and keeps searching internal documents until it has enough information. Employees in department B ask the same assistant the same question. If the application doesn’t scope searches to each user, the department B business plan can end up in an answer to an employee in department A. If you replace the departments with customer companies, the same problem occurs in a multi-tenant software as a service (SaaS) architecture where multiple customers share one assistant and knowledge base.
#Amazon Bedrock Managed Knowledge Base#Amazon Bedrock AgentCore Gateway#Amazon Bedrock AgentCore#Model Context Protocol#Amazon Cognito