Authorised access
Limit agent sessions to the projects and repositories explicitly authorised for each developer.
BatonX is an enterprise security and governance platform for AI coding agents, designed to run in your own infrastructure.
It helps organisations bring authorised repository access, isolated workspaces, centrally managed policies and auditable agent activity into the developer workflow.
Designed for customer-managed deployment
BatonX Control Layer
Governance overview
Organisation controls
Approved repositories · Central policies · Audit visibility
Repository scope
Explicitly authorised
Actions & access
Policy governed
AI coding agent session
Starting with Claude Code
The challenge
AI coding agents can read files, execute commands and interact with external services. Introducing them into an enterprise requires clear boundaries around what they can access, which actions they can take and how their activity is recorded.
BatonX is being built to give engineering and security teams a shared way to define and enforce those boundaries, while preserving the native coding-agent experience as much as possible.
A governance layer designed to fit into developer workflows — with controls and operational visibility for the organisation.
Platform capabilities
Bring access, runtime boundaries, policy and visibility into one governance approach for AI-assisted development.
Limit agent sessions to the projects and repositories explicitly authorised for each developer.
Constrain agent filesystem access to the approved workspace through technically enforced runtime boundaries.
Manage organisation and repository instructions alongside controls for commands, network access and sensitive operations.
Centrally manage approved model connections and protect provider credentials from direct exposure wherever technically possible.
Give authorised teams visibility into session activity, policy decisions, security events, usage and cost.
Deployment model
BatonX is designed for customer-managed deployment, giving your organisation control over its installation, configuration and operational responsibilities.
An important distinction. On-premises deployment does not automatically mean that model processing stays on-premises. Where model requests are processed depends on the approved provider or endpoint configured for your deployment.
Your environment
Customer-managed installation
BatonX configuration
Organisation-owned setup
Access & policy controls
Repository and session boundaries
Coding agent workspace
Approved project context
Integration direction
The initial integration focuses on Claude Code. The architecture is designed to support additional coding agents and model endpoints over time, with a consistent governance foundation.
Support for each integration will be confirmed as it becomes available.
The first integration focus for BatonX’s governance approach to AI-assisted coding.
A consistent governance foundation intended to accommodate broader support as integrations become available.
Product by CreatySoft
BatonX brings CreatySoft’s enterprise software engineering focus to the adoption of AI coding agents.
Our approach centres on practical developer workflows, technically enforced controls and clear operational ownership.
Practical workflows
Built around how developers work.
Enforced controls
Boundaries designed to be technical.
Clear ownership
Operational responsibility stays visible.
Talk to us about your development environment, security requirements and potential pilot scope.
BatonX · A product by CreatySoft