Show HN: We are fixing memories for AI Teams

supercontext.dev

Hi HN!

We built SuperContext because we got tired of re-explaining our codebases to AI tools every single time. Now our whole team can jump between Claude Code, Cursor, and Windsurf without losing context.

It's open source and uses MCP to connect with your existing tools. Would love to hear what you think!

Open articleView on HN

Hey Harsh, cool stuff! Can you elaborate on what's the difference than me just passing in my repo's context to cursor? Is it aggregating all of my team's KBs such as Slack, Notion, etc?

that's exactly the goal - aggregating context from slack, notion, and other services. but for now, it works a bit differently than just passing your repo context to cursor. right now, multiple team members can access the shared memory that your AI agents are generating. so when your colleague has a conversation with claude code about some architecture decision, you can pick up right where they left off when you switch to cursor or windsurf. the context persists across different AI tools through MCP servers, rather than everyone starting fresh each time. think of it as a shared brain for your team's AI development tools.

ahh okay! happy to chat more in this and and help you w feedback if you'd want

Reply on news.ycombinator.com