Anthropic on Thursday introduced the Skills Update Suite, a feature that trains Claude on tasks tailored to the way you work. This update shows how the market is moving beyond the large language model itself and focusing more on use cases.
As part of the update, AI Model Maker released the Agent Skill Specification as an open standard. Agent skills extend your ability to: claude user According to Anthropic, the goal is to create, deploy, share, and discover new skills for agent AI.
Vendors are adding organization-wide management capabilities that allow company administrators on Team and Enterprise plans to manage skills in a central location. Users also have access to pre-built skills from Anthropic partners such as Canva, Notion, Figma, and Atlassian.
The move to make agent skills an open standard follows Anthropic’s success with the Model Context Protocol, an open standard for large-scale language models to communicate. The vendor donated MCP to the Linux Foundation earlier this month.
Dig deep into your skills
The Skills for Claude expansion comes two months after Anthropic introduced Agent Skills, a new way to build specialized agents using files and folders. Users can add files within folders to describe processes, such as filling out forms or browsing websites, to teach agents how to perform specific tasks.
“Many business use cases that require human input can now actually use skilled AI models to perform those actions on your behalf,” said Gartner analyst Arun Chandrasekaran. “These models will be given more agent capabilities, allowing them to perform actions on behalf of the user.”
Skills also allow agents to run software without interacting with another tool or tools. Coordinate between different programssaid Lian Jye Su, an analyst at Omdia, a division of Informa TechTarget.
“Agents can solve tasks themselves, in which case they become smarter and much easier to deploy,” Su said. “Obviously, this will require a lot of skills, relevant management systems and updates.”
With the newly expanded skill suite, Anthropic aims to create a skill library and a set of predefined skills that can be applied to different types of tasks, Chandrasekaran said.
concern and maturity
Giving agents additional capabilities means giving them more autonomy to perform certain tasks, which can raise concerns among businesses.
But that’s why companies need to have strong AI security and governance strategies, Su said.
“Companies need There are guardrails, filters, and firewalls. When you have all the different agents with the skills, they will be in the right place,” he said.
Anthropic’s skill development efforts are an example of how the AI market’s focus has shifted from model updates to use cases.
“The industry is now moving beyond models to asking, ‘What can a model do for me? How can a model orchestrate a workflow? How can a model redesign a process? How can AI become autonomous within our environment?'” Chandrasekaran said. Apart from anthropology, he added: OpenAI operator They are similar to skills because agents perform specific tasks. “These are all kind of incremental features that we need to move toward our goals.”
