Custom AI Agents

Agents that work on your own data.

Write the instructions, select the contact, company, location, web and column data the agent may use, then run it across every row of your list.

Start for free Explore the 3-minute demo →
What you get

One agent, defined by you, applied to everything

Not a fixed list of AI features — a layer you configure for your own motion.

Your own instructions
You write what the agent should do. It is not limited to a preset menu of tasks.
You choose the context
Pick what the agent may use: a person, a contact, a company, a location, available web data, your table columns and the signals you collected.
Runs across a list
Define an agent once and apply it to every row, instead of repeating the same prompt by hand.
Writes into a column
Agent output lands in a column of your table, so it sits next to the rest of the record and can be exported.
Research and qualification
Company research, pain points, personas, buying power, competitors, technologies and open roles are all instruction-level choices.
Personalisation
Generate an opener or a tailored value proposition from what the agent found, rather than from a generic template.
3-step process

How does it work?

Define once, apply to the whole list.

Write the instruction
Describe the task in your own words — what to look for, what to conclude, what format you want back.
Choose the data it may use
Select the context: contact and company fields, location, available web data, existing columns, collected signals.
Run it on your rows
Apply the agent to the list. Results are written into a column, ready to review, filter and export.
Use cases

What teams build agents for

Two recurring jobs that stop being manual.

Pre-call research
You have a call in an hour
Instead of opening six tabs per account, an agent gathers what matters — company context, likely pain points, technologies, open roles — into one column you can read at a glance.
See how →
List qualification
Two hundred rows, unequal quality
An agent applies your own qualification criteria to every row and writes its conclusion in a column, so you sort a long list without reading it one by one.
See how →
Related features

Works well with

AI Research Web and company research from your workspace. AI Summaries Ready-made prospect summaries. AI Personalization Openers written from real context. AI Enrichment Complete columns your providers do not cover.
See all AI & Automation features →

Custom AI agents, answered

What you control and what runs automatically.

Do agents run on their own without me?

No. You define the agent and choose when to run it and on which rows. It is not an autonomous process running in the background.

What data can an agent see?

Only what you give it: the contact, company, location, table columns, collected signals and the web data available in your workspace.

Where do the results go?

Into a column of your table, alongside the record. From there you can filter, review and export them like any other field.

Do agents consume credits?

Yes. AI generation consumes credits, and the amount varies by model. The published cost table is on the pricing page.

Can I reuse an agent?

Yes. Once an agent is defined it can be run again on other rows and other lists.

Ready to build your first agent?

Start for free. No credit card required.

Start for free Book a demo
This site is registered on wpml.org as a development site. Switch to a production site key to remove this banner.