Your stories, your vision. AI-powered execution.

Frequently Asked Questions

How does Velocity work with AI coding assistants?

Velocity provides a native Model Context Protocol (MCP) integration. AI assistants like Claude Code can connect directly to your project, discover prioritized work, claim stories, and track their progress through tasks.

The key benefit is session continuity—agents can resume exactly where they left off, with full context of what they were working on, what tasks are complete, and any comments or feedback from the team.

Can I still use Velocity without AI agents?

Absolutely. Velocity is a full-featured project tracker that works great for human teams. The velocity-based planning, automatic iteration management, and story workflow are valuable whether you're working with AI or not.

The AI integration is an enhancement, not a requirement. You can adopt it gradually as your workflow evolves.

What is "velocity-based planning"?

Just like Pivotal Tracker, Velocity measures how many story points your team completes per iteration (a 3-iteration rolling average). It uses this to automatically fill your backlog into future iterations—you can see at a glance when features will likely ship based on actual throughput.

This approach respects your team's natural pace and makes it easy to communicate timelines without constant manual recalculation.

Can I import my data from Pivotal Tracker?

Yes. Export your project from Pivotal Tracker as a CSV file, then use our import tool to bring over your stories, labels, epics, and comments. The import preserves your story states, estimates, and history so you can pick up right where you left off.

We also support full CSV export from Velocity, so your data is never locked in.

Is there a free trial?

Yes! You can try Velocity free during our open beta. This gives you full access to all features including MCP integration so you can see how it works with your team and AI assistants.

How do I connect Claude Code to my project?

Simply point Claude Code at your Velocity MCP endpoint. You'll be taken through a quick authentication flow, and within moments Claude Code will have access to your projects and can start working on stories.

What happens if an AI agent produces work I don't like?

You review all delivered work before accepting it. If something isn't right, you can reject the story with feedback explaining what needs to change. The rejection reason is stored on the story, so when the agent picks it up again, it has full context of your feedback.

This creates a natural feedback loop where you stay in control of quality while the AI handles implementation.