MCP · bring your own agent

Connect your AI agent to your discovery tree

Outcomify speaks the Model Context Protocol, so the AI you already use — Claude or any MCP client — can read, search, and propose on your Opportunity Solution Trees. Point your agent at your research; it drafts the opportunities and solutions, you approve them.

Add it to Claude

$ claude mcp add --transport http outcomify https://api.outcomify.com/mcp

Or point any MCP client at https://api.outcomify.com/mcp over the HTTP transport.

Get started freeMeet Canopy AI

What your agent can do

The server exposes your discovery work as tools an agent can call — safely scoped to your access.

Read your tree

List organizations and trees, read and search nodes by title, type, status, or tag, and pull the full history of what changed.

Propose, don’t clobber

Your agent can draft new opportunities and edits as proposals a human reviews and commits — or write directly when you want it to.

Shape the structure

Create outcomes, opportunities, and solutions; set outcome metrics and health; tag, move, and reorder nodes.

Bring the evidence

Attach interviews, metrics, feedback, and sources to the nodes they support, so the tree stays grounded in what you heard.

Work your feedback board

Ask what customers want that nobody has picked up, read a request with the needs behind it, and, with write access, approve, reply, move, and merge requests. Replies are shown to you before they go out.

Read your funnels

List your key funnels, see where people drop off, and follow each step week by week against what shipped — every change with its margin of error. Link a step to the outcome, opportunity, or solution it measures.

Connect in three steps

1

Add the server

One command for Claude (or point any MCP client at the HTTP endpoint).

2

Authorize

Sign in with Outcomify. Your agent acts as you, with your access — nothing more.

3

Point it at a tree

Ask it to list your trees, then work inside one — read, search, and propose against it.

Try asking your agent

  • “Read my “Q3 activation” tree and propose three opportunities from these interview notes.”
  • “Find every opportunity tagged “Bug” and summarise the supporting evidence.”
  • “Turn this screenshot of our Miro tree into proposed nodes in Outcomify.”
  • “What do customers ask for on our feedback board that nobody has picked up?”
  • “Did the invite prompt we shipped move activation?”

Canopy, or your own agent — your call

Canopy (built in)

No setup. Outcomify’s own AI drafts opportunities and solutions you review and approve, right in the app.

Your agent (MCP)

Bring the AI you already use. It works across your other tools in one conversation, and proposes to your tree with the same approve-first safety.

Frequently Asked Questions

Connecting your AI agent to Outcomify over MCP

It’s a Model Context Protocol (MCP) server that lets an AI agent — Claude, or any MCP-compatible client — connect to your Outcomify account and work with your Opportunity Solution Trees and your feedback board: reading and searching nodes, proposing opportunities and solutions, setting outcome metrics, attaching evidence, reading and triaging customer requests, and reading your funnels. MCP is an open standard for connecting AI assistants to tools and data.

Canopy is Outcomify’s built-in AI — no setup, it drafts opportunities and solutions you review and approve inside the app. The MCP server is “bring your own agent”: you connect the AI you already use (Claude, etc.), which can work across your other tools in the same conversation. Both follow the same safety model — agents propose, you approve — and the MCP server can also write directly when you ask it to.

Run: claude mcp add --transport http outcomify https://api.outcomify.com/mcp — then sign in with Outcomify when prompted. Any MCP client that speaks the HTTP transport can connect to https://api.outcomify.com/mcp the same way.

Your agent authenticates as you and only sees the organizations you can access. For changes it can propose drafts that a human reviews and commits, mirroring how Canopy works; direct writes happen only when you ask for them. You can disconnect the server at any time.

The MCP server is part of Outcomify — sign up free to connect your agent. See pricing for plan limits.

Keep exploring

Canopy AI

Outcomify’s built-in AI assistant — draft and approve, no setup.

AI for Opportunity Solution Trees

How AI fits into continuous discovery without taking over the thinking.

Product overview

The OST canvas your agent reads and proposes against.

Point your agent at your discovery tree

Connect Claude or any MCP client to Outcomify and let it propose opportunities and solutions — grounded in your research, approved by you.

Start FreeMeet Canopy AI