The Forecast MCP server is now available. It connects your Forecast data to AI assistants, so instead of clicking through Forecast to find answers or make updates, you can just ask.
Use it through Ask AI, built right into Forecast, or connect the AI assistant your team already uses.
Jump to...
What's New at a Glance
- Two ways to use it: Ask AI in Forecast works with nothing to set up, or you can connect your preferred AI assistant, such as Claude, ChatGPT, or Microsoft Copilot Studio.
- Get answers in plain language: Ask about project and portfolio health, team capacity, time and billing, and holiday conflicts.
- Take action through conversation: Reassign work, move tasks through your workflow, reschedule tasks, phases, and allocations, and log time.
- Stay in control: Every change is previewed first, and nothing is applied until you confirm it.
Feature Details
Ask AI, built into Forecast
Click Ask AI in the top-right corner of Forecast and describe what you need in your own words. The assistant works out the rest.
Connect your own AI assistant
Because MCP (Model Context Protocol) is an open standard, the Forecast MCP server isn't tied to a single AI tool. If your organization already uses a compatible assistant, you can add Forecast to your existing setup instead of adopting something new. Supported assistants include Claude, ChatGPT, Perplexity, Grok, Mistral Le Chat, and Microsoft Copilot Studio.
You can also combine Forecast data with other sources connected to your assistant, like email or accounting systems, in a single conversation.
Ask questions about your work
Get a quick read on how things are going without building a report or opening multiple screens.
Try asking:
- "Which projects are at risk right now?"
- "Who has availability next month?"
- "How much non-billable time was logged on the (Project Name) project this quarter?"
Take action in a single request
Handle updates that would normally mean opening task after task.
Try asking:
- "Reassign all of (Team Member Name)'s tasks on the (Project Name) project to (Team Member Name)"
- "Push the (Phase Name) phase on the (Project Name) project back by 2 weeks"
- "Log 90 minutes to the (Task Name) task"
Built-in safeguards
Every request uses your signed-in Forecast permissions, so the same access rules you have in the application always apply.
Before any update is made, the assistant shows you exactly what will change and waits for your approval.
When making several changes at once, you can deselect anything you want to skip. You'll also be alerted if a change would push work past a project's end date or clash with another allocation.
Where to go next
- Forecast MCP Server: An introduction to Forecast MCP and what it can do for you.
- Forecast AI: Getting Started: Start using Ask AI, or follow the setup steps to connect your preferred AI assistant.
- Forecast AI: Capabilities and Prompt Examples: Explore everything you can ask and do, with example prompts.
- How Forecast's AI Features Use Your Data: Learn how your data is handled.
Comments
0 comments
Please sign in to leave a comment.