Stop opening Meta Ads Manager.

Meta shipped a connector for Claude and ChatGPT on April 29 that runs write operations directly on your Facebook ad account, no API keys, no app review.

Okay the title is half clickbait, but the substance behind it is the part nobody is covering correctly, so here goes.

First, understand what Meta actually shipped. It's two surfaces. There's the Meta Ads MCP server, which is what you point Claude or ChatGPT at to control your ad account by talking to it. And there's a Meta Ads CLI, which is the same capability surfaced as a command line tool for engineering teams building internal pipelines. Both are in open beta, both are free, and both authenticate through a normal Meta login. That last part is the unlock honestly, because every prior attempt at agentic Meta ads required going through the Marketing API, which means dev credentials, app review, and business verification. The MCP server collapses that whole credential dance into clicking "log in with Meta."

The capability itself is split into four buckets and they actually let you write, not just read. Reporting is the obvious one, pull performance, surface insights, the standard stuff. Campaign management is where it gets interesting because the agent can create and edit ads, ad sets, and full campaigns by natural language. Catalog management lets it build product catalogs and fix data feed problems. Signal diagnostics covers Conversions API health, learning phase status, and signal prioritization. Most third-party ad-MCP wrappers (AdKit shipped one for Google plus Meta on May 1, two days after Meta's own launch) are read-only, because writing to ad accounts through unofficial APIs is risky and Meta gates it hard. Meta's own connector doesn't have that limitation because Meta is the platform.

Now the part where I'm not just selling it. It's open beta, so the rate limits and edge cases aren't fully documented. The bigger thing to know is the learning phase, which is roughly 50 optimization events that Meta uses to actually figure out who to show your ad to. If your agent edits a campaign mid-learning, the campaign resets and your performance tanks. Meta hasn't built guardrails for this yet, which means the agent has zero idea by default. You have to tell it. If you don't, you'll watch your CPA spike on the first day you let Claude touch your account and blame the connector when it was actually you. (fyi this is the same thing that happens to humans editing campaigns mid-learning, the AI just does it faster.)

Okay. The connector is interesting. The actual move is what you build on top of it.

Because here's the thing. If you point one Claude window at your ad account and ask it to "manage my ads," you get a slightly faster Ads Manager. That's not the win. The win is splitting the work across three agents, each with a job, each with its own context.

Agent one is research. It pulls competitor angles, audience signals, top-performing creative patterns, the stuff a strategist would normally Slack you about on a Tuesday. Output is a brief.

Agent two is mission control. It reads the brief, makes the edits, respects the learning phase, and stops itself when it hits a decision that needs you. It's the one that actually touches the account.

Agent three is diagnostics. It watches signal health, Conversions API events, and the learning phase counter. If something drifts, it pages the other two, or you. Always on, runs in the background, doesn't ask permission to look.

Three agents, one ad account, one prompt window if you want, or three separate ones if you'd rather keep contexts clean. That's mission control. That's the thing the dashboard literally cannot give you, because the dashboard is a UI, not a workflow.

Who this actually changes things for. If you're a business owner running your own ads, you stop opening Ads Manager. You ask the research agent what's working, you tell mission control to ship the changes, you let diagnostics yell when something breaks. If you're an agency, the diagnostics agent is basically a junior account manager that costs you in tokens instead of salary. One ad account per context, repeat the stack across clients, scale linearly without hiring linearly. If you're a builder, the CLI is the better surface because you're not using this through chat, you're wiring it into scheduled jobs, internal dashboards, slack alerts. Same backend, different glove.

Personally, I haven't run this end to end yet, the connector dropped four days ago. The architecture I'm planning on setting up is exactly the three-agent split above, Claude on top, Opus 4.7 for mission control because the edits are the highest-stakes calls, Sonnet 4.6 on diagnostics for the always-on cost, and probably Sonnet on research since that's mostly retrieval and summarizing.

If you're still treating Meta ads as something you log into a dashboard to manage because that's how it's always been, this is the moment to look up. The dashboard isn't going away, but it's not the default surface anymore. The default is an agent, and the real question is whether you're the one building the stack, or the one paying somebody else's wrapper for less capability.

https://www.facebook.com/business/help/1456422242197840

7 comments