Anthropic hands Agent Skills to the Linux Foundation's Agentic AI Foundation
Anthropic is transferring the Agent Skills standard, its trademark, GitHub organisation and domain to the Agentic AI Foundation, following its earlier donation of the Model Context Protocol. The standard is used by more than 70 agent platforms and has around one million published skills with nearly 280 million installs.
Anthropic has donated Agent Skills, the open standard it created for packaging specialised knowledge and workflows for AI agents, to the Agentic AI Foundation, which operates under the Linux Foundation. The move mirrors the company's earlier transfer of the Model Context Protocol and is intended to put a widely adopted format under neutral governance.
A skill is deliberately simple: a folder containing a SKILL.md file that describes a task or domain, together with any scripts or reference material the agent needs. The format emerged from Anthropic's work on Claude Code and was published as a standard in December 2025. Since then it has been adopted by more than 70 agent platforms including ChatGPT, Cursor, GitHub Copilot and VS Code, with roughly a million skills published and close to 280 million installations.
The donation includes the Agent Skills name and logo, the GitHub organisation and the agentskills.io domain. Engineers from Anthropic, Google, OpenAI and other companies will maintain the specification in their individual capacities, with David Soria Parra and Mahesh Murag as lead maintainers. Anyone can use or contribute to the standard without joining the foundation, which Anthropic, Block and OpenAI co-founded in 2025.
With MCP covering how agents connect to tools and data, and Skills covering how they acquire procedural knowledge, the foundation now stewards two of the main interoperability layers of the agent ecosystem.
Why it matters
Standards under neutral governance reduce the risk that enterprises build on one vendor's format only to find it abandoned or forked. For agent builders, it means skills written once can travel between platforms, which lowers switching costs and encourages investment in reusable institutional knowledge.
Indicops view
We already package client-specific procedures, from a state's scheme-eligibility rules to a bank's KYC exception handling, as skills so that they can be reviewed, versioned and reused across agents. Neutral governance of the format strengthens the case for Indian government departments to maintain their own libraries of approved skills, which is a far more durable asset than any single chatbot.
More in Agents
Back to the newsroomCognition launches SWE-2 Flash in Devin at one-fifth the price of SWE-2
Cognition has released SWE-2 Flash, a faster and cheaper coding model available inside Devin Desktop and Devin CLI. The company says it costs about a fifth of SWE-2 and runs three times as fast, and is free for paid Devin users until 30 November, though its weights remain gated.
Anthropic previews Claude Duet, a proactive agent that works from its own virtual machine
Anthropic has opened a waitlist for Claude Duet, a research preview in which Claude runs on a dedicated virtual machine, monitors connected apps and messages the user only when something needs attention. Built on Claude Opus 5.5, it ships as a standalone app for US Pro, Max and Ultra subscribers with hard limits on financial actions.
Frontier labs converge on agentic workflows as the next platform layer
Across OpenAI, Anthropic and Google, the centre of gravity has shifted from chat to agents that plan, call tools and complete multi-step work. Enterprise buyers are now evaluating orchestration, observability and guardrails as first-class requirements.