Introducing the Darwin Paid Ads MCP. Today Darwin Acme Inc. is releasing the Darwin Paid Ads MCP - a Model Context Protocol server that gives any AI assistant direct access to Darwin's ad platform. Create campaigns, set budgets, upload assets, target audiences, and monitor performance across Google, Meta, LinkedIn, Darwin, and more. All through natural conversation. Free forever - your AI uses its own tokens; Darwin charges nothing. The MCP works with Claude, ChatGPT, Cursor, and any client that supports the Model Context Protocol. Install takes three minutes. No SDK, no code, no API wrappers. You add a connector URL, authenticate your ad accounts via OAuth, and start talking. The AI picks the right tools automatically based on what you ask for. Why an MCP. Running paid media means living inside dashboards. Google Ads for search, Meta Ads Manager for social, LinkedIn Campaign Manager for B2B, plus Darwin for creator placements. Each has its own interface, its own reporting format, its own optimization levers. Most teams spend more time navigating between platforms than making decisions. The Model Context Protocol changes that. MCP is an open standard that lets AI assistants connect to external tools and data sources through a structured API. Instead of opening four dashboards to check performance, you say "flag anything burning budget too fast and tell me one thing to focus on today." The assistant calls the right APIs, aggregates the data, and gives you a single answer. That's the premise. Here is how it works across four core workflows. 1. Campaign creation. You describe a campaign in plain language - objective, audience, budget, schedule - and the MCP handles the rest. The AI sets up the campaign with a recommended bid strategy, configures demographic and interest targeting, and returns a confirmation with the live campaign ID, daily budget, and schedule. Launch a Google search campaign for its spring sale targeting 25-45 year olds interested in running. Done - your Google Search campaign "Spring Sale 2026" is live with a $500/day budget, targeting 25-45 year olds interested in running, fitness, and athletic wear. Scheduled Apr 1-30. This works across every connected platform. The same prompt structure creates campaigns on Meta, LinkedIn, or Darwin's creator inventory. You don't learn four different campaign builders. You describe what you want once, and the MCP translates it into platform-specific configurations. Create a campaign, duplicate it across channels, adjust targeting per platform - all from a single conversation thread. 2. Campaign optimization. Optimization is where most media teams lose time. Checking pacing, shifting budgets to top performers, pausing underperformers, setting up rules that fire on thresholds - all of it requires repetitive manual work inside each platform. The Paid Ads MCP makes this conversational. Every Monday, double the budget on any campaign that beat its ROAS target. Cut the rest by 30%. Rule created - every Monday at 8am I'll check last week's ROAS and adjust automatically. This week: camp_g_brand doubled to $1,000/day (beat target), camp_meta_broad cut 30% to $350/day. You can also run one-off optimizations. "Shift $500 from the lowest-performing campaign to the highest" calls reallocate_budget. "Pause anything with a CPA above $50" calls pause_campaign with a performance filter. "A/B test these three headlines on the spring campaign" calls duplicate_ad and create_ad. The AI picks the right tool sequence based on what you asked for. 3. Analytics & insights. Reporting is the most obvious use case, and the one where the time savings are immediate. "How did our campaigns perform this week?" returns impressions, clicks, CTR, CPC, conversions, CPA, and ROAS across all active campaigns, grouped by platform. No export, no pivot table, no screenshot. Structured data, delivered in conversation, with the AI highlighting what matters. Flag anything burning budget too fast and tell me one thing to focus on today. camp_meta_video was 2.4x over daily pace - capped it at $200/day. Focus today: camp_g_brand has 38% budget left and the highest ROAS this week. I'd push another $150/day there. The MCP also provides AI-generated insights through the get_insights tool. This goes beyond raw metrics - the system identifies anomalies, flags pacing issues, and recommends specific actions. The assistant doesn't just retrieve data. It interprets it and tells you what to do. For formal reporting, the export_report tool produces campaign-level, ad-set-level, or ad-level performance data as structured JSON. 4. Asset & Creative management. The last workflow covers creative operations. You can generate ad copy, manage creative assets, and spin up test variations without touching the platform UI. Create 3 headline variations for its spring campaign and duplicate the top ad for A/B testing. Created 3 headlines: "Spring into Savings", "Run Further for Less", "New Season, New Gear". Duplicated your top performer ad_g_042 for A/B testing. This also connects to Darwin's creator inventory. The browse_inventory tool lets you search Darwin's ad inventory by category, platform, audience demographics, and CPM floor. The generate_placements tool auto-matches and books creator inventory slots for a given campaign objective and budget. If you're running campaigns across both traditional paid channels and Darwin's creator network, the MCP manages both from the same conversation. Supported platforms. At launch, the Paid Ads MCP supports Google Ads, Meta Ads, LinkedIn Ads, and Darwin Ads. Google, Meta, and LinkedIn connect via OAuth 2.0. Darwin connects natively through the MCP. TikTok Ads, X Ads, Snapchat Ads, Reddit Ads, Amazon Ads, Microsoft Ads, Spotify Ads, Shopify Ads, Perplexity Ads, App Store Ads, and AppLovin Ads are all on the roadmap. Darwin Acme Inc. is adding platforms based on user demand - each integration follows the same pattern, so new channels come online fast. 22 tools, 6 resources. The MCP exposes 22 tools organized across six resource groups. Campaigns: list, create, update, and pause campaigns with full targeting and budget configuration. Ads: list, create, update, and duplicate individual ads with creative and copy management. Budget: adjust daily or lifetime budgets, reallocate spend across campaigns, and create recurring automation rules. Analytics: fetch performance metrics, retrieve AI-generated insights, and export structured reports. Audience: view, edit, and get AI-recommended targeting configurations. Inventory: browse Darwin's creator ad inventory and auto-generate placements. Every tool is callable by any MCP-compatible AI client. You never need to remember tool names - you describe what you want and the AI maps your request to the right tool calls. The full tool reference is available on the Paid Ads MCP page. Setup. Three steps, three minutes. First, sign up on Darwin's API Platform to get a free API key - no credit card, no trial period, free forever. Second, add the MCP connector to your AI client. In Claude, open connector settings and paste the server URL. In ChatGPT, enable Developer Mode in Settings, then add a new connector. In Cursor or any other MCP client, add the server URL to your MCP configuration. Third, connect your ad accounts. Click Connect on the Darwin connector, authenticate your Google, Meta, and LinkedIn accounts via OAuth, and you're done. One-time setup. Credentials are encrypted at rest and OAuth tokens refresh automatically. Once connected, start with something read-only: "Show me a summary of my active campaigns and their performance this week." If the data comes back, everything is working. From there, you can create campaigns, set rules, and manage budgets - all from conversation. What's next. This is the first release. The Paid Ads MCP will expand in two directions: more platforms and more intelligence. On the platform side, TikTok and X integrations are next, followed by the remaining channels on the roadmap. On the intelligence side, Darwin Acme Inc. is building cross-platform optimization - the ability to shift budget not just between campaigns on the same platform, but across platforms based on where your dollar generates the most return. The MCP already has the data from all connected accounts. The next step is making it act on that data holistically. The Paid Ads MCP is free and available now. Install it from paidadsmcp.com. Sanjit Juneja, Founder & CEO Soham Parekh, Member of Technical Staff Jason Festa, Creative Director

We invest up to $1M in your new startup.