Skip to content

AI agents that fill your pipeline.

Tell Clu the job. Agents find your buyers, research accounts, keep your CRM clean and report the week.

Watch Clu work.

One sentence in, an agent out. Then ask Clu from any Slack channel.

Meet Clu, the GTM agent in Cleanlist AI
Meet Clu0:54
Clu in Slack: ask your GTM agent from any channel
Clu in Slack0:36

An agent for every GTM job.

Start from one of 12 or describe your own. Each runs on a schedule or the moment your CRM changes.

The Cleanlist AI Agents library with 12 agents, grouped as Prospecting, CRM hygiene, Research and Reporting. Each card opens that agent's page: its home, its instructions and connections, its trigger, the steps of its play (each step opens its settings), its runs with Clu's review, and its settings. The CRM duplicate cleanup agent opens on its duplicate review, where groups of duplicate Salesforce records wait for a person to approve them before they merge.

See every agent, how to build it and what it costs

Watch Clu build an agent.

Clu builds agents that prospect, enrich, run outreach and keep your CRM clean.

Who to find, how many a week and where they should go, in one sentence.

  • 41,500 people match before you spend a credit. At 500 a week, that's 83 weeks of new names.
  • Your ICP is already loaded. Clu reads your company profile, so the titles, industries and company sizes are filled in.

Which titles count, what “new” means, which day it runs and how much it can spend per person.

  • Pick or type. Every question comes with answers ready.
  • Every answer becomes a rule. Clu writes it into the agent's instructions as you go.

Find people, get emails and phone numbers, verify them, add them to the sequence, post to Slack. Open any step to change it.

  • Every step shows its filters and its price. Email and phone cost up to 11 credits a person, charged only for what's found.
  • Edit it in plain English. Say “skip anyone already in a sequence” and Clu changes the play for you to accept.

The test lists who it would add, with their verified emails. It spends no credits and writes nothing to Salesforce.

  • 24 of 25 came back with a verified work email. 21 had a phone number too. You see every name before anyone is added.
  • One yes turns it on. Clu asks once: turn it on for Monday at 8:00 AM?

It finds the week's new people, enriches and verifies them, starts the sequence and posts the run in #pipeline.

  • Clu reviews every run. The first: 500 found, 487 verified work emails, 412 phone numbers, all 500 in Series B outbound.
  • The summary lands in Slack. In the channel you picked, with a link to every lead.
Warp: Cold call pickups went from 8% to 13%.

An interactive demo of Clu in the Cleanlist AI app. Alex types one sentence on Home: Every week, find 500 new people who match my ICP, get their emails and phones, and add them to Series B outbound. Clu makes the agent Weekly ICP outbound, reads the company profile and finds 41,500 people who match the ICP. It asks four questions and builds the play (new people who match the ICP, email, phone and profile, verify, add to Series B outbound, post to Slack). Alex asks it to skip anyone already in a sequence; Clu proposes the change on the canvas and Alex accepts it. A test run on real data spends no credits and writes nothing. Then Alex turns it on: the first Monday run finds 500 new people, 487 with a verified work email and 412 with a phone number, adds all 500 to the sequence and posts to #pipeline. The buttons above the app build two more jobs: a quarterly CRM duplicate cleanup and inbound routing. The app's screens and tabs can be clicked.

Search or ask Clu⌘K
IntegrationsReports
Clear
New chatEnrich my latest list
Alex Kim
12,340 creditsAdd
Hey Alex, ready to get started?
Find VPs of Marketing at Series B SaaS companies
Owner Favorites
NameLeadsCompaniesLast openedCreated
Inbound, September1,040688YesterdaySep 1, 2026
Q4 target accounts2,400180Sep 28, 2026Sep 22, 2026
Webinar attendees316241Sep 24, 2026Sep 18, 2026
Series A founders1,2001,050Sep 9, 2026Aug 30, 2026

Every lead researched and scored.

Agents read the company site, the careers page and the team page, with the source on every answer.

Skills running on one inbound lead, Ines Delacroix, VP of Sales at Haldrevo AI. The Qualify inbound leads agent's Research skill reads the company website, its careers page and her bio on its team page and writes a brief with a source on every line. Its Qualification skill then scores her 88 out of 100, qualified, with four reasons. The browser's tabs, the source chips and each reason can be opened.

Ines DelacroixVP of Sales · Haldrevo AI Qualify inbound leads
haldrevo.example/team/ines-delacroix
Ines DelacroixVP of Sales at Haldrevo AI · New York

Experience

  • VP of SalesHaldrevo AI · 4 mos
  • Head of SalesFenvolta · 3 yrs 2 mos
Research skillCompany and contact 3 sources · 21s5/record
  • What they doAI agents for customer support, sold to enterprise support teams.
  • HiringTwo Account Executive roles and an SDR role open, in New York and San Francisco.
  • The personJoined as VP of Sales 4 months ago, after 3 years leading sales at Fenvolta.
  • Why call nowA new sales leader hiring her first reps. They need pipeline from week one.
Qualification skillAgainst your ICPQualified5/record

Who to call first, after every run.

Clu posts the review in Slack, with a button for each decision that's yours.

The Runs tab of the Qualify inbound leads agent, and the review it posts in the inbound channel in Slack when a run ends. The newest run scored 2 inbound leads: Naomi Osei at Distyl scored 95, call her first, and Sofia Marchetti at WellSaid scored 58. The filters and the run links work.

Agents Qualify inbound leadsLast 30 days
Runs
Records
Credits
Credits per run
StatusStartedDurationRecordsCredits
Slack # inbound

Qualify inbound leadsAPP

2 new inbound leads. Call Naomi first.

  • Naomi Osei Distyl95Call first
  • Sofia Marchetti WellSaid58Nurture

Naomi fits your ICP on title, size and stage. 20 credits used.

Distyl is an account Tom owns in Salesforce. Send Naomi to him?

Watch an agent work.

Clu enriches your customer champions and drafts a personal reconnect for each owner.

Customer champion agent: reconnect with champions at their new company
Customer champion agent0:46

10,000+ users, from founders to full sales teams

  • Switched fromZoomInfo
  • Case study
  • HackerOneSwitched fromApollo
  • Switched fromClay
  • Case study
  • Sumo LogicSwitched fromClay
  • Switched fromApollo
  • Case study
  • Switched fromZoomInfo
  • Switched fromZoomInfo

Questions before you turn one on.

How credits work
What does a Cleanlist AI agent do?

An agent does one sales job on its own, on a schedule or the moment a CRM record changes. Example: every Monday at 8:00 AM, find 500 new people who match your ICP, get their emails and phones, and add them to Series B outbound.

How is Clu different from an agent?

Clu is the assistant you talk to, in Cleanlist AI and in Slack. An agent is one standing job Clu sets up for you. Clu builds it, edits it and reviews every run.

Do I have to build agents myself?

No. Tell Clu the job in one sentence and it drafts the whole agent, credit limit included. Test it on real data, then turn it on.

How do I know an agent got it right?

Every run shows its records, the credits it used and Clu's review. Research answers cite a source on every line, and you test each agent on real data before it goes live.

What does a run cost?

Credits per lead: 5 to qualify, 1 per verified work email, 10 per phone number, 0 to search. A run stops at the credit limit you set.

Where do results and updates go?

Results go to a list, then to HubSpot or Salesforce when the list syncs. Clu sends each run's review to a Slack channel, a DM or your inbox, with a link to every lead.

Build an agent today. It runs Monday.

Describe the job in one sentence. Clu builds the agent and you turn it on.

Try Pro free for 14 days: 250 credits, 3 seats.