Agent experience
Interesting to see SaaS companies starting to report MCP usage on earnings calls:
- Datadog: MCP tool calls up 4x q/q, 22x since Q4 2025
- Figma: MCP write usage up 75% q/q
- Atlassian: MCP calls up 400% q/q
Sean Roberts
@JavaSquip
- There it is! wondered when this was going to come full circle. “We need this as an industry solve” seems like a good idea. With them, it’s Vercel top of funnel in a tench coat. “Sign up for Vercel” happens quick!! couldn’t keep the logo off the “oss” agent plugins social imgSkill𝑠𝑒𝑡𝑠
- @Netlify has this shipped in a few mins thanks to a clear spec and aligning on existing conventions!Build a plugin once and use it across compatible agent clients. Introducing Agent Plugins, an open standard developed with @awsdevelopers, @cursor_ai, @github, @code, and @vercel that packages Agent Skills and supports MCP server configurations in a shared format.
- All of y'all coming to my neck of the woods for @RenderATL need to stop by my workshop on axis.run to eval the quality of your AX. If nothing else you can look at this guy's goofy faces for 25 minsMost teams shipped something for agents this year with no way to know whether it worked. At @RenderATL, @JavaSquip closes that gap live: he'll score a real project with AXIS, the open-source framework for measuring agent experience, and separate model problems from the slow
- If there's a benchmark, you can now optimize for it - trivially (without cheating or even tradeoffs as long as the tradeoff space is covered). Being able to accurately reproduce what you want to improve in a lab setting has never been more useful (or critical).
