Agensi is a marketplace for AI agent skills, modular, installable capability packages that extend what AI coding and productivity agents can do. Think of it as an app store, but instead of apps for your phone, it sells skills for your AI agents like Claude Code, OpenAI's Codex CLI, GitHub Copilot in VS Code, Cursor, Gemini CLI, and over 20 other supported tools.
The core concept is refreshingly simple. Skills are built on an open standard called SKILL.md, a structured file format that any compatible agent can discover and use automatically. Buyers purchase a skill once (no subscriptions or license keys), download a zip file, drop it into their agent's skills folder, and it just works. The agent picks it up on the next session without any configuration.
The marketplace hosts both free and paid skills spanning categories like code review, database migration auditing, SEO optimization, content generation, Git commit writing, changelog generation, and more. Every listing goes through a security review process before it goes live: automated scanning for dangerous command patterns, hardcoded secrets, obfuscated code, and prompt injection, followed by admin review. It's a curated catalog, not a scrape of random GitHub repos.
For creators, Agensi offers an 80/20 revenue split (creators keep 80%), buyer-fingerprinted downloads for IP protection, and a piracy reporting dashboard that can trace leaked files back to the exact buyer. There's also a Skill Request Board where potential buyers post what they need with upvotes and bounties, so creators can build to proven demand rather than guessing.
One of the more interesting features is that the catalog itself is machine-readable, meaning AI agents can query it mid-conversation. So when a user's agent encounters a task it doesn't have a skill for, it can surface the right Agensi listing at exactly the moment the user needs it.
Agensi positions itself at the intersection of two trends: the rapid adoption of AI coding agents and the emerging ecosystem of extensible agent capabilities. It's betting that as agents become the primary interface for development and productivity work, there will be a thriving market for specialized, plug-and-play skills, and that a trusted, curated marketplace will win over the alternative of hunting through scattered open-source repos.
