AWS Bedrock
Use AWS Bedrock models in engini with your Bedrock key and Region: chat, invoke, images, embeddings, guardrails, and as the model behind an AI agent step.
Book a demoAWS Bedrock for AI agents and workflows.
engini calls the foundation models you use on AWS Bedrock, in the Region you choose, with your own Bedrock key: chat completions, direct model invocation, image generation, text embeddings, token counts and guardrails. Bedrock can also be the model behind an AI agent step in a workflow, so the agent reasons with a model in your AWS account while engini governs what it may do in your systems.
Cursor, your code
audit log
in engini
What a workflow can do with AWS Bedrock. Every tool is one you allow or leave out.
The AWS Bedrock tools in engini, for calling your models from a workflow step or from your code.
Model
Get a reply from a model.
Images and embeddings.
Check content against a Bedrock guardrail.
Count tokens before you send.
Catalog
See what your key can use in the Region.
Any endpoint
Call any Bedrock endpoint, with the key added for you.
Triggers
There are no Bedrock triggers. A model step runs inside a workflow started by any other trigger.
Connect your AI agent to AWS Bedrock. Four ways in, one set of permissions.
Claude, ChatGPT, Cursor
Create an MCP server in engini with your AWS Bedrock connection and only the tools you choose. Add its address to your client and sign in through engini. There is no key to copy.
MCP client setup →Python and TypeScript
Hand your agent a toolset of AWS Bedrock tools. The SDK gives OpenAI and Anthropic models the tools in the format they expect and runs the ones they pick.
SDK guide →For coding agents
The engini CLI connects AWS Bedrock and runs its tools from a terminal, with machine-readable output, so an AI coding agent can drive it without getting stuck.
CLI guide →Any language
Every AWS Bedrock tool is an endpoint. Call it with exact inputs from a scheduled job or a sync, with no model in the loop.
API reference →.cursor/mcp.json. Claude Code: claude mcp add.// .cursor/mcp.json: Cursor, connected to AWS Bedrock through engini
{
"mcpServers": {
"engini-aws-bedrock": {
"url": "<your engini MCP server address>"
}
}
}
Create the MCP server in engini with your AWS Bedrock connection and the tools this agent may use, then paste its address. The client signs in through engini on first use. Claude, VS Code and other clients take the same address: see the MCP client setup guide.
Each request runs through the AWS Bedrock tools in the agent's toolset or MCP server. If a tool is not allowed, the agent cannot call it.
What teams build on AWS Bedrock. Agents and workflows, across systems.
Documents read inside your AWS account
Invoices and contracts are read by your Bedrock model, and the workflow writes the result to the ERP.
An agent step on your own model
An AI agent step uses Bedrock as its model and works through the tools you give it in SAP or Salesforce.
Replies checked by a guardrail
Drafted replies pass a Bedrock guardrail before they reach Zendesk.
Embeddings for your documents
Policy documents are embedded for search by your own agents.
Every integration runs both ways, so any of these flows can also run in reverse. These are examples, not the limit: any flow between AWS Bedrock and another system engini connects is available.
You decide what each agent may do in AWS Bedrock. engini enforces it on every call.
Only the tools you allow
A toolset lists the AWS Bedrock tools an agent may call and the connection each one uses. Give one agent read tools only and another the write it needs. engini refuses any call outside the toolset.
Scoped when you set them up
An MCP server exposes only the connections and tools chosen when it is created, and engini shows who uses each server.
Whose permissions apply
Connect a person's own AWS Bedrock account and engini acts as that person, with their AWS Bedrock permissions. Use a shared integration account and it acts as that account. Every run is recorded under the user who started it.
Kept out of prompts
Your agent never needs your credentials: engini stores them and adds them when it runs each call, so keys stay out of prompts and tool inputs.
Every call on record
Every call lands in History and the audit log, which you can export or stream to your SIEM.
Test before production
Workspaces keep teams apart, and development, staging and production environments keep a test agent away from live data.
Connect AWS Bedrock in three steps. Then decide what each agent may do.
Sign in to AWS Bedrock
Create a Bedrock API key in AWS that may use the models you need, then add an AWS Bedrock connection with the key and your Region.
Choose the tools
Add the connection to an MCP server or a toolset with only the tools each agent needs.
Build or ask
Use the connection as a step in a workflow, or ask your agent to work in AWS Bedrock from Claude, ChatGPT, Cursor or your own code.
AWS Bedrock and engini. Straight answers.
Can an AI agent write to AWS Bedrock, or only read?
Both. a Bedrock step returns text, images, embeddings or a guardrail result; what it can change in your systems depends on the workflow's steps and the tools you give the agent. Whether a given agent may use them is up to you: give it a toolset or an MCP server with read tools only, and it cannot call the rest.
Whose AWS account is used?
Yours. The connection holds your AWS Bedrock API key, engini adds it to each call, and usage is billed by AWS to your account.
Can the model change data in my systems?
Only through the tools you give it. A model step in a workflow returns text or data; what is written to your ERP or CRM is decided by the workflow's steps or by the toolset of the agent, and every call is recorded.
Does the agent act as me in AWS Bedrock?
Model calls use your Bedrock key, in your Region. Every run is recorded under the person who started it.
What if the tool I need is not listed?
Use Send API Request to call any AWS Bedrock API endpoint through the same connection, and give it only to the agents that need it.