We use cookies to understand how you use our site and improve your experience. Cookie policy
Last updated:
Palette works as an MCP (Model Context Protocol) connector inside your AI tools. Add it once, and every conversation starts with your org's context already loaded.
You open Claude, ask about your company, and it already knows who's on the team, what they're working on, what decisions were made, what's blocked. No pasting. No "let me give you some context." It just knows.
Palette appears as an MCP connector in your AI tools, similar to how you'd connect Slack, Google Drive, or any other app. Once added, your AI tool can pull your org's context on demand.
In Claude (chat, code, or cowork), Palette shows up as a connected integration. Claude can read your org context, ask about team priorities, surface recent signals, and ground its answers in what's actually happening across your org.
In ChatGPT, Palette connects as an MCP connector in settings. Same context, same experience. Ask about your org and get real answers.
When Palette is connected, your AI tools have access to:
This isn't raw data from your tools. It's interpreted, structured, plain-English context that's been confidence-scored and traced to source.
You could plug Slack, Linear, and Notion straight into your AI tool. But then your AI only knows what you have access to — your channels, your projects, your docs. You've given your AI the same silo you're in. It still misses what other teams are working on, deciding, and prioritizing.
When every team connects their tools to Palette, something different happens. Your AI gets context from tools and channels you're not in — because those teams opted in. Engineering's Linear, Marketing's Notion, Sales's Slack channels — all mapped, structured, and available as context for everyone's AI.
Every team that connects makes every other team's AI smarter. That's not a feature — it's a network effect.
Palette also does the interpretation upfront. Your AI gets pre-structured context: who's here, what matters, what changed. Smaller payload, better answers. Your AI gets the answer, not the homework. See how Palette works for the full flow.