The team that has the what-if ready.
The Planner Agents are the FP&Agents for planning and modeling: a growing team of autonomous agents that keep the forward view current and the model logic in your team's hands.
The problem
The plan is stale before it ships.
By the time the reforecast lands, the business has already moved, and the what-if that mattered died in the queue.
Reforecasts take weeks
Assembling actuals and rebuilding assumptions turns a forecast refresh into a project.
What-ifs die in the queue
Every scenario is a model rebuild, so the questions that deserve answers never get them.
Logic changes need a specialist
When changing a formula feels dangerous, the model stops matching the business.
What the Planner Agents do
Every plan current, every what-if ready.
The Planner Agents are Cube's FP&Agents for planning and modeling: a growing team of autonomous agents that build the forward view, stress-test scenarios, and change model logic from plain language. They work on your clean, decision-ready data, so every forecast traces to the source.
Planner Agent
Describe the change in plain language; the plan updates.
Smart Forecaster
Predictions from your history, seasonality, and drivers, with assumptions you control.
Formula Modeler
Builds and edits model logic in plain language.
Every agent here works under Charlie, the super agent: one question in, one decision-ready answer back.
Planners, in product
One plan, managed end to end.
The plan the agents keep current: live metrics, assumptions tagged AI or User, and an action log that always ends at your review.
The payoff
Planning gets its week back.
Reforecasts happen in the flow
New actuals land and the plan updates around them. No rebuild week, no starting over.
You describe, it drafts
Say the change in plain English. The model updates the drivers and formulas for you, using your own numbers.
You hold the publish button
Nothing goes live until you review it, and you can open any figure to see where it came from.
MCP writeback
Reforecast in any LLM. Published back to Cube.
Ask for the reforecast in any LLM. The Cube MCP Server reads your actuals, and the new plan version writes back to Cube, where it waits on your approval.
+ any endpoint
Reforecast FY26 revenue off actuals: 8% growth, headcount flat. Publish it as a preview.
Cube MCP Server
Your AI reads it, and writes it back.
The Cube MCP Server gives Claude, ChatGPT, Copilot, and the agents you build read and write access to the same clean, decision-ready data the Planner Agents work on.
Read: decision-ready answers
Pull the current forecast, scenario deltas, and driver assumptions into Claude, ChatGPT, or your own planning agent, live and traced.
Write: work that lands in Cube
Publish forecast versions, scenario assumptions, and driver changes back to the model, through approvals, permissions, and the audit trail.
Use cases
Where the Planner Agents go to work.
Rolling forecasts
Refreshed with the latest actuals the moment assumptions change.
Scenario planning
Downside, base, and upside from one driver change, honestly recalculated.
Headcount planning
People plans and financial targets kept in the same conversation.
See it on your own workflow.
Pick the scenario your board keeps asking about and watch the Planner Agents run it end to end.
Ask for the what-if out loud.
Book a demo and run your hardest scenario through the Planner Agents in a live sandbox.
Part of the FP&Agents: one super agent, four specialist teams.
