We use cookies to understand how you use our site and improve your experience. Cookie policy
Last updated:
Here's how to get started with the Palette context layer.
Already using Palette Desktop? The context layer is built in. You still need to connect your tools (step 1) and let the context layer build (step 2), but your Desktop agents read it automatically — you can skip the MCP install in step 3.
You choose which tools to connect and which channels, projects, or spaces to include. Currently supported:
You stay in control of what Palette can see. No bulk data dumps, no access you didn't grant.
Once connected, the context layer synthesizes signals from your tools into a living context map of your organization. This covers who's here, what they're working on, how teams are structured, and what's in focus right now.
Depending on how much data is in your connected tools, this can take up to 30 minutes. The result is a structured, plain-English representation of your org that stays current as things change.
Add the Palette MCP server to the AI tools your team already uses:
It's a one-time setup per tool. See the guides for , , and .
Your AI tools now begin every session with your org's real state. They know who's on the team, what's in flight, how decisions get made. No more copy-pasting context into every prompt.
For example prompts and workflow patterns, see Context to your AI and Context to your workflows.
We're currently onboarding design partners. Learn more and apply.