
An open protocol for connecting AI models to tools and data sources.
MCP began when two Anthropic engineers proposed a build-once protocol to end rebuilding every integration from scratch. Shopify's CEO and Microsoft's Kevin Scott championed it, and it became widely adopted. Lesson: open up integrations as a protocol, and expose every product feature through it, since Claude could not act on projects that weren't exposed. No usage numbers are given.
From Dianne Penn on Anthropic Labs
MCP came out of Anthropic Labs, which chases big bets outside the core roadmap. Anthropic Labs incubated MCP alongside Claude Code, Skills and Claude Design. Dianne Penn, who leads product for Anthropic's research and Labs teams, helped incubate it.
Labs looks for bets that could be worth 10x, 100x or 1000x, and the first question for any idea is whether "there is a there there". Prototypes are expected to teach the team something even when nothing ships.
Run ambiguous bets like MCP with tiny pods and loosely held prototypes. Labs keeps pods small, sometimes a single engineer, because large teams on vague, big ideas slow down. It hires people who want zero-to-one work and is willing to turn bets off, then revisit them one or two model generations later.
Penn: "you can be very strongly held opinion about the theme or the area and then more weekly held about the exact prototype." Turning off a bet is hard because each owner acts like a founder who has put heart and soul into it.
Came out of Anthropic Labs' incubation work alongside Claude Code, Skills and Claude Design.
MCP came out of Anthropic Labs, which chases big bets outside the core roadmap. Anthropic Labs incubated MCP alongside Claude Code, Skills and Claude Design. Dianne Penn, who leads product for Anthropic's research and Labs teams, helped incubate it.
Labs looks for bets that could be worth 10x, 100x or 1000x, and the first question for any idea is whether "there is a there there". Prototypes are expected to teach the team something even when nothing ships.
Run ambiguous bets like MCP with tiny pods and loosely held prototypes. Labs keeps pods small, sometimes a single engineer, because large teams on vague, big ideas slow down. It hires people who want zero-to-one work and is willing to turn bets off, then revisit them one or two model generations later.
Penn: "you can be very strongly held opinion about the theme or the area and then more weekly held about the exact prototype." Turning off a bet is hard because each owner acts like a founder who has put heart and soul into it.
Started because every integration was rebuilt from scratch. Two engineers proposed a build-once protocol, and Shopify and Microsoft's Kevin Scott became early champions. Krieger wants every product feature exposed through it so Claude can act on them.
Developer toolsA chip company whose GPUs and CUDA software power gaming, AI and accelerated computing.
Developer toolsAnthropic's agentic coding tool that lets Claude work on code end to end.
Developer toolsAn AI code editor that lets engineers build software with AI while staying in control.
Developer toolsA code hosting and collaboration platform for software developers, with free and paid plans.
Developer toolsAn AI app builder that lets people create software by chatting with an agent.
An app development company that builds apps for founders and clients.
Developer toolsA fast, well-designed project and issue tracking tool for software teams.
Developer toolsGoogle's backend platform for building apps.