Unified AI Skill Format
The Unified AI Skill Format is a standardized specification for packaging and deploying discrete AI capabilities across different organizational platforms and agent systems. Developed through collaboration between Anthropic, OpenAI, and Microsoft, the format establishes common protocols for defining AI skills—reusable units of AI functionality—enabling interoperability between AI agents and organizational infrastructure. By creating a shared technical standard, the format reduces friction in integrating AI capabilities from multiple providers into unified agent ecosystems.
Technical Specification
The format defines how AI skills are structured, documented, and made discoverable to agent systems. It includes specifications for skill metadata, interface definitions, capability descriptions, and execution requirements. This allows AI agents from different platforms to recognize, request, and execute skills regardless of their original developer or underlying implementation, similar to how software APIs enable cross-platform functionality.
Organizational Adoption
The format supports the broader shift toward agent-first organizational infrastructure by providing a common language for skill exchange and integration. Organizations can build flexible AI systems that combine skills from multiple vendors without creating vendor lock-in or requiring significant custom integration work. This encourages a modular approach to enterprise AI deployment, where specialized capabilities can be assembled and reconfigured to meet changing business needs.
Source Notes
- 2026-04-07: Anthropic, OpenAI, and Microsoft Just Agreed on One File