344
Productivity & Workflow355
Automation & Workflow225
Software Development251
Marketing & Growth193
AI Infrastructure & MLOps176
Writing & Content Creation204
Data & Analytics142
Photography & Imaging156
Design & Creative170
Customer Support133
Sales & Outreach125
Education & Learning131
Voice & Speech135
Operations & Admin87
Figma announced 20 new partners for MCP, a shared way for AI tools to connect to other apps, while CEO Dylan Field said the standard has flaws.
In short: Figma announced 20 new partners for MCP, even as its CEO said MCP is not a perfect way for AI tools to connect to outside apps and data.
Figma, a popular design software company, said it is expanding support for MCP, short for model context protocol. MCP is a shared set of rules that helps AI agents (software that can take actions for you, like an assistant) connect to other tools and data sources. You can think of it like using the same type of plug so different devices can connect.
At Figma’s first Config conference in India, the company announced 20 new MCP partners. The list includes Amp, ClickUp, Dia, Junie by JetBrains, Linear, Meta Muse, Mistral, Notion, OpenCode, and Sarvam Agents.
The announcement comes after public discussion about how open Figma should be with its data through MCP. Dylan Field, Figma’s co-founder and CEO, has said he wants Figma to support many standards, but he also believes MCP has “tons of room for improvement.” He also raised a concern about “asymmetric context sharing,” which is a situation where one side can get more access to information than the other.
Field said Figma is not trying to expose all of its data through MCP because doing so could increase the risk of abuse. He added that most people would not use the access in harmful ways, but the company still has to plan for that possibility.
As more apps add AI helpers, standards like MCP can make it easier for tools to work together. But this also raises a basic question for users, which is how much access an AI helper should get to your files and work history, and what safeguards should be in place.
Source: TechCrunch AI