GitHub: Copilot Cloud Agent Can Now Be Selectively Enabled Per Organization
Why it matters
GitHub has enabled enterprise administrators to selectively activate access to the Copilot cloud agent through custom properties, replacing the previous all-or-nothing approach. The new feature brings more granular control over AI agent capabilities at the level of individual organizations, with new API endpoints and management through the AI Controls interface within GitHub Enterprise settings.
GitHub has announced a significant upgrade to access management for the Copilot cloud agent — an AI assistant that can autonomously work on coding tasks in the cloud. Enterprise administrators can now selectively control which teams and organizations have access to the cloud agent, replacing the previous all-or-nothing model.
What Do Custom Properties Bring to Access Control?
Previously, enterprise administrators could only globally enable or disable access to the Copilot cloud agent for an entire organization. The new feature introduces custom properties — configurable attributes that administrators can set at the level of individual organizations within an enterprise.
This means a company with 50 development teams can enable cloud agent access only for teams working on approved projects, while keeping it disabled for teams working with sensitive data or regulated systems. Control is exercised through the AI Controls interface in GitHub Enterprise settings.
What New API Endpoints Are Available?
Alongside the visual interface, GitHub has also published new API endpoints that enable programmatic access management. This is essential for organizations that automate infrastructure management — DevOps teams can integrate cloud agent access control into their existing automation pipelines.
The API supports reading and setting custom properties, checking access status per organization, and bulk operations for configuring multiple organizations simultaneously.
Why Is Granular Control Important?
Cloud agents that autonomously write and execute code represent a significant security vector. The ability to precisely control who has access to these capabilities is critical for enterprise users who must balance productivity against regulatory requirements and internal security policy.
This upgrade reflects a broader industry trend: AI tools are becoming powerful enough to demand the same level of access control that organizations apply to production infrastructure.
This article was generated using artificial intelligence from primary sources.
Related news
Anthropic and NEC build Japan's largest AI engineering workforce — Claude for 30,000 NEC employees
AWS: multimodal biological foundation models accelerate drug discovery by 50 percent and diagnostics by 90 percent
CNCF: infrastructure engineer migrated 60+ Kubernetes resources in 30 minutes with the help of an AI agent