$ loading_
Securely let AI agents perform privileged actions with short-lived governed tokens.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "Broker" yet — see the docs or source repo.
Use Broker to mint a short-lived token only for creating a GitHub issue, then create an issue in my-org/my-repo titled "Fix login timeout issue" with reproduction steps and expected behavior. Keep an audit trail.
A policy-governed execution result, such as confirmation that the issue was created with related audit details.
I need an AI agent to perform a privileged action. Use Broker to mint a one-time short-lived token with least privilege, limited to this specific action, and report whether it passes policy checks.
A short-lived single-purpose authorization result, including whether policy enforcement allows the action.
Use Broker to perform a controlled action and output an audit summary including the action purpose, permission scope, and execution result.
An operation result summary with audit information for traceability and review.
When developers or product managers want an AI agent to perform actions such as creating GitHub issues, Broker can mint short-lived single-purpose tokens to reduce the risk of exposing long-lived credentials.
In workflows that require privileged automation, teams can use Broker to constrain action scope with policy enforcement so only validated requests are executed.
For AI agent actions that require accountability, Broker provides audit logging to help teams record what controlled actions were performed and within what scope.
Broker is an MCP tool that lets AI agents securely perform privileged actions. It mints short-lived single-purpose tokens on demand with policy enforcement and audit logging.
Based on the description, Broker uses short-lived single-purpose tokens instead of long-lived credentials. This narrows the permission scope and adds security through policy checks and audit logging.
The provided material does not include installation steps, runtime requirements, or credential setup details. Please see the source repository for specifics.
Run SSH and database aliases safely without exposing passwords to AI.
Securely authorize AI agents to sign up for SaaS dev tools within budget limits.
Run AI agents with fine-grained access control and zero-setup brokering.
Securely connect AI IDEs to AWS Bedrock AgentCore and discover tools.
Issue short-lived GitHub tokens for AI agents to securely access repos and APIs.
Add governance and security controls for AI coding agents on GitHub.