How Teams Share AI Agent Skills

Greg Isenberg32:27
1 VIEW
0 comments · 0 votesOpen discussion
Sign in to join the discussion

    Video summary

    Greg Isenberg and Remy Gaskell describe agent skills as reusable operating proceduresAn AI agent skill is a reusable package of instructions, resources, and tool guidance for performing a bounded kind of work. that package instructions, examples and resources for a repeatable task. A central GitHub repository gives a team one controlled sourceAn AI agent skill registry is a maintained catalog of reusable, versioned capabilities available to AI agents. while plugins expose those skills to tools such as Claude Code and CodexAn agent plugin extends an AI agent with packaged tools, integrations, skills, or interfaces that are not part of the agent's base capability..

    Greg Isenberg and Remy Gaskell separate personal experiments from approved team skillsTeam skill governance defines how shared agent skills are proposed, reviewed, approved, maintained, and retired., then use version control and update checks to distribute improvementsAgent skill versioning records changes to a reusable skill so teams can review, distribute, compare, and restore specific versions. without silently replacing local work. They also show how usage views and linked skill chains can reveal which procedures matter and where a workflow needs refinement.

    Original YouTube thumbnailWatch on YouTube