Meta Ads AI Connectors Add Granular Security Controls for AI Agents.

Meta Ads AI Connectors now include additional security settings for working with Claude and other AI agents. Advertisers can define more precisely what AI is allowed to do inside individual ad accounts.

Ads AI Connectors allow AI agents such as Claude to interact with Ads Manager through MCP. Meta has also expanded the range of available MCP tools significantly.

In May, Ads AI Connectors offered 29 MCP tools. Now, connector settings in Claude provide permission controls for nearly 100 MCP tools.

The ability to create campaigns, ad sets, and ads, adjust budgets, change targeting, and modify other settings was already available. The key update is more granular control over these actions for individual ad accounts.

In Meta Business Suite > Settings > Integrations > Ads MCP server, users can see the ad accounts and catalogs available through MCP.

For each ad account, seven actions can now be managed separately:

• change budgets;
• create campaigns;
• create ad sets;
• create ads;
• change targeting;
• change creatives;
• change status.

For example, Claude can be prevented from creating campaigns, changing budgets, or editing targeting in a specific ad account.

There is also more control during the Ads AI Connector setup process. Users can now choose which Facebook Pages and Instagram accounts the connector can access. However, selecting specific ad accounts directly during the connection process is still not available.

Additional restrictions can also be configured on the Claude side. Individual MCP tools can be allowed to run without confirmation, blocked completely, or set to require approval before execution.

The Ads MCP server is still not available to all accounts.

Our Telegram Channel - @SpyLeader

For Spy Service please write - @TrafficSponsor