Meerkats AI

Welcome to Meerkats

What Meerkats does, how it fits together and where to start.

Meerkats is the context layer that lets your AI agents work reliably in your business.

Meerkats connects to the tools you run your business on and keeps their data current in one place. It defines each metric once, so every agent and every report uses the same number.

Agents work on top of that business context. They check your metrics on a schedule and follow the rules you set. When a rule matches, Meerkats proposes a change, for example pausing a Google Ads campaign or lowering a Meta budget. The proposal waits in your Inbox, and a person approves it before it runs, unless you let that rule act on its own. Agents can also send you reports and alerts by email or Slack. Meerkats records every run.

You can bring your own AI tools too. Claude connects to Meerkats over MCP, and your own code can use the Meerkats API. Both read the same business context.

Meerkats does not run your ads, and it does not replace your ad platforms, store or CRM. It changes nothing in a connected tool unless a rule allows it. Each workspace is kept separate, and your data is never used to train AI models.

How Meerkats fits together#

  1. ConnectionsYour ad platforms, marketplaces, store, CRM and Slack, kept in sync.
  2. Business contextMetrics defined once, current state and history, per workspace.
  3. Agents and rulesCheck your metrics on a schedule and decide what needs attention.
  4. InboxProposed changes wait here. A person approves each one, unless you allow it to run alone.
  5. ActionsThe approved change is made in the connected tool.
  6. Run historyEvery check, proposal and change is recorded in Agents run history.

What you can connect#

In every tool you connect, Meerkats can create, read, update and delete records. What it can do in each one depends on the permissions you grant and the permissions the tool allows.

ToolWhat Meerkats works with
Meta Adscampaigns, ad sets, ads and their performance
Google Adscampaigns, ad groups, ads, keywords, search terms and their performance
Amazon AdsSponsored Ads performance, plus sales from Seller Central
Flipkartad campaigns, wallet and performance, plus product revenue
Shopifycustomers, orders, products and inventory
Microsoft Advertisingcampaigns and their performance
TikTok Adscampaigns and their performance
Walmart Connectretail media campaigns and their performance
Blinkitads and sales from the Brand Portal
Zeptoads and sales from Brand Central
Swiggy Instamartbrand ads and sales
Razorpaypayments and refunds
GA4site behaviour and conversions
HubSpotcontacts, companies, deals and pipeline
Salesforceleads, accounts and opportunities
Apollopeople and company data
Slackmessages in the channels you choose
Gmailemail in the inboxes you connect
DocumentsSOPs and playbooks you upload
Data warehouseyour own tables in BigQuery, Snowflake or Postgres
Competitor trackingcompetitor prices and reviews on Amazon, Flipkart, Zepto and Blinkit
Inventorystock levels by product
APIdata from any other system you run

Where to go next#

Last updated October 1, 2026