What happened
Anthropic announced the donation of the Agent Skills standard, including its name, logo, GitHub organization, and domain, to the Agentic AI Foundation (AAIF). This move places the standard under the Linux Foundation, where it will be maintained by individual engineers from various companies rather than solely by Anthropic.
Anthropic has donated the Agent Skills open standard to the Agentic AI Foundation (AAIF), a project under the Linux Foundation. The donation includes the Agent Skills name, logo, the associated GitHub organization, and the agentskills.io domain. This transfer is designed to ensure the standard remains open, community-driven, and vendor-neutral.
Agent Skills is a standard that provides AI agents with specialized knowledge and workflows through folders containing a SKILL.md file and supporting scripts. Developed during the creation of Claude Code, the standard addresses the gap between general knowledge and the procedural knowledge required for real business tasks. Since its publication in December 2025, more than 70 agent platforms, including ChatGPT, Cursor, GitHub Copilot, and VS Code, have integrated the standard.
The ecosystem has seen significant growth, with one million skills published and nearly 280 million installed. Following the initial launch in October 2025, developers quickly created loaders for other tools like Codex, Cursor, and OpenCode, demonstrating broad industry interest in standardizing agent capabilities.
Maintenance of the standard will now be handled by individual engineers from Anthropic, Google, OpenAI, and other companies. David Soria Parra and Mahesh Murag have been designated as Lead Maintainers. The AAIF will provide a neutral home for the project, ensuring that the community has a say in its future direction, similar to how the Model Context Protocol (MCP) is governed.
Source details: anthropic.com ↗
Why it matters
This transition ensures the long-term vendor-neutrality and community governance of a critical interoperability standard for AI agents. By moving Agent Skills to the AAIF, similar to the earlier donation of the Model Context Protocol (MCP), the industry reduces the risk of a single company controlling the evolution of agent capabilities. This supports broader adoption across competing platforms like ChatGPT, Cursor, and GitHub Copilot, fostering a more open and standardized ecosystem for agentic workflows.
The donation of Agent Skills to the AAIF mirrors the earlier transfer of the Model Context Protocol (MCP) to the same foundation. This pattern suggests a strategic shift in the AI industry toward neutral governance for critical infrastructure standards, reducing vendor lock-in and encouraging cross-platform compatibility.
With over 70 platforms already integrating the standard, the move to a neutral foundation is crucial for maintaining trust among developers and enterprises. It ensures that the evolution of agent skills is not dictated by a single commercial entity, which is essential for a standard that underpins how AI agents interact with tools and data.
The standard's focus on procedural knowledge fills a significant gap in current capabilities. By standardizing how this knowledge is packaged and shared, the AAIF can facilitate a more robust and secure ecosystem for agentic AI, allowing agents to perform complex, domain-specific tasks reliably across different software environments.
Interactive Mechanism: How It Actually Works
Explore the underlying technology behind this development interactively.
crm_get_transaction(id='4092').An agent must create a draft calendar event for Tuesday at 2 p.m. Which evidence would establish the requested result?
What to watch next
Monitor the implementation of the proposed roadmap, which includes versioned releases, a formal proposal process, and security guidance for skill developers. Watch for how the new Lead Maintainers, David Soria Parra and Mahesh Murag, manage community input and specification changes through the AAIF's transparent decision-making processes.
The proposed roadmap includes versioned releases, a formal proposal process, and security guidance for skill writers and installers. The implementation of these elements will determine how effectively the community can contribute to and secure the standard.
Community engagement through the project’s Discord and GitHub discussions will be key to shaping the specification. The transparency of the decision-making process under the AAIF will be a critical factor in maintaining the standard's credibility and adoption rate.
The involvement of engineers from multiple major AI companies as individual maintainers will test the feasibility of cross-company collaboration on a shared standard. Success in this model could serve as a template for other AI infrastructure projects seeking neutral governance.