I’m excited to announce the beta of Freckle's CLI — our biggest release since we first launched!
Freckle’s CLI is the Clay-alternative that lets you build GTM workflows straight from the terminal. We took the core primitives of our existing product and rebuilt them to be agent-native: waterfall enrichments, research agents, a workflow engine, signals, and GTM tech stack integrations.
Our goal with the CLI is to enable agents to build GTM workflows using natural language, faster than a human could in Freckle or Clay today. And importantly, at the same or better quality.
By doing so, we could evolve the role of our users from manually building tables and workflows to directing agents to do this work at scale, much like how software engineering has evolved.
Two weeks ago, we hit this benchmark and started using our CLI to build 100% of our GTM workflows internally. Now there’s no turning back.
Today, I wanted to give you a preview of the beta and how we got here.
You can join the beta waitlist here: https://freckle.io/cli. We’re opening up more slots each week!
Freckle’s mission has always been to help GTM teams get better data without complexity.
Over the past 2 years we’ve grown to more than 7,000 users, who have primarily come to Freckle for the ease of use and scalability of our SaaS product.
But during the last six months, our customers have started asking for something different: the ability to build enrichment workflows directly from coding agents (Claude Code, Codex, etc.).
What started as a niche few feature requests quickly became a consistent theme on customer calls. Users didn’t want to be leaving the terminal to go to a product’s UI if it could be avoided. So we quickly leaned into building for this way of working.
What immediately felt evident was that this would shift what GTM software needs to be. The product surface is no longer just the UI — it’s the API, the schemas, the provenance, and the observability around every action an agent takes. And the UI itself would likely need to change to match what a user needs when operating from a coding agent vs. operating from the UI itself.
This presented a massive opportunity. If we got this right, we could deepen our differentiation around ease of use and scalability — and meet the best GTM operators where they’re now working.
So instead of dipping our toe in the water, we went all in. We dedicated almost all of our resources to building Freckle around that future: callable by default, observable by design, and simple enough that humans and agents can use the same GTM data layer.
After many iterations, and a few dead ends, we landed on an approach that is working at scale and feels magical to use.
With Freckle’s CLI, you can state exactly what you’re looking to accomplish in natural language to Claude Code, and the agent will figure out the enrichment and orchestration logic for you. This could be something like:

If the agent doesn’t have the context it needs from your repo, it may ask some qualifying questions about how you define your ICP and your scoring criteria.

Then it will look at all the available data providers, integrations, skills, and pre-built logic Freckle has to devise the optimal workflow for your use-case.

We wanted this to feel materially better than using Clay’s natural language builder, Sculptor. Sculptor misses the mark because a human can build a Clay table faster and with better logic than Sculptor can. It was designed more to teach someone how to use Clay, not to do the work itself.
In contrast, Freckle’s CLI encodes all the logic to build high-quality workflows faster than a human, and has been designed to be the default way of using Freckle moving forward.
It’s quotes like these from our design partners that give me confidence we’re onto something special:

Our public launch of Freckle’s CLI is coming soon, and in the meantime the waitlist for beta access is here: https://freckle.io/cli. We’re releasing more beta slots each week and can’t wait for you to experience it.
If you’re an existing customer, please reach out to Emily (emily@freckle.io) or myself and we will be sure to help you transition your tables and workflows if you’d like to use the CLI.
I want to also give a massive shout out to the team who have worked hard to bring this to market so quickly. It’s a special feeling to see a startup in full-flight, building a product that unlocks a new way of working. The energy of this team is something I am always grateful to be a part of.
Here’s to our next chapter!
Nathan