New Delhi: Anthropic recently announced its next step toward that goal, Skills for Claude. The tool is made up of folders that include instructions, scripts, and resources that Claude can load when needed to make it smarter at specific work tasks, such as working with Excel and following their organisation’s brand guidelines. People can also build their own skills for Claude relative to their particular jobs and use them across the Claude AI, Claude Code, Anthropic’s API, and the Claude Agent SDK. Box, Rakuten, and Canva have already used the tool.
The feature is designed to improve Claude’s AI agent capabilities for their work specifically, so users don’t have to spend as much time writing the perfect prompt or referring to past context every time users try to accomplish tasks, and it is available to Pro, Max, Team, and Enterprise users. Anthropic layer on top of Claude’s PowerPoint Skill, they have to create a presentation about how Haiku 4.5 is doing in the market. OpenAI, Anthropic, Google, Microsoft, and others have been working toward the goal of actually helpful AI agents for years.
Executives regularly bring up agentic AI on earnings calls and redirect internal resources toward building the tools. Progress has mainly been incremental, with the companies fighting to release the latest feature updates or iterations of agents. OpenAI unveiled AgentKit, a group of tools that executives stated were designed to help users take agents from the prototype to production and targeted both big companies and individual developers. OpenAI also announced a consumer-facing tool that allows people to work with applications inside ChatGPT, like Zillow and Uber Eats.
