Upsolve AI
Day 4 after signup
🐳 Your data agent can now write to Notion
Attach any MCP server so the agent can save analyses, open tickets and pull context.
Why it works
- Open with a concise summary that states the new capability and its practical value before getting into details.
- Arrange the explanation as clear steps, then reinforce it with a short list of controls and safeguards.
- Offer distinct next steps for existing customers and newcomers so each audience can act appropriately.
How it is built
- Newsletter series greeting
- Feature summary
- Problem and benefit
- Setup instructions
- Capabilities overview
- Try-it calls to action
Email copy
Hey there, Welcome back to your Upsolve AI newsletter! The first two issues covered what your agent knows and who can see what. This one is about what your agent can do with an answer once it has one. TL;DR Attach any MCP server to your Upsolve agent, such as Notion, Linear, GitHub or PostHog. The agent still answers from governed warehouse data, and can now read and write where your team works. You choose which tools each server exposes and tell the agent when to use it. 🔌 Plug Notion, Linear and more into your data agent. A good analysis usually gets copied by hand into a doc or a ticket, and it loses its context on the way. Useful context also sits in tools your data agent can't reach, and every new integration used to mean custom engineering work. 🛠️ How it works Add a server. Paste the MCP server URL in project settings and add a one-line instruction, like "use this to save analyses in Notion docs". Authenticate and choose tools. Connect through OAuth, then allow all tools or block the ones you don't want the agent to use. Ask and act. Ask a data question, then tell the agent to write the analysis to Notion. It finds the right page and creates it. ▶ Watch Serguei's full demo 📦 What's included Any MCP server: Notion, Linear, GitHub, PostHog and any other tool that exposes an MCP server. Per-tool control: allow or block individual tools on every server you connect. Instructions per source: tell the agent when and why to use each connection. Still governed: questions still run through your data model and row-level security. 🚀 Try it Already a customer? MCP Connections are live in your workspace. Add the server your team writes in most, then ask the agent to save its next analysis there. New to Upsolve? Start free at upsolve.ai → with 2,000 AI credits, or read the full feature page. Next issue: Harness 2.0, a data agent that's twice as fast and learns every user. Questions? Reply to this email or reach us at kaling@upsolve.ai Follow the series on YouTube → Subscribe to Upsolve AI Building the future of data-intelligent AI, The Upsolve Team The conduit between data and decisions Stay tuned for more exciting updates, customer stories, and news from us! - Team Upsolve AI You are receiving this email because you opted in to receive email from the list Product Updates & Releases from Upsolve AI Upsolve AI, 1111B S Governors Ave STE 6587 Dover, DE 19904 Unsubscribe