
Security News
Frontier AI Is Now Critical Infrastructure
The Fable shutdown shows how quickly model access can become a business continuity risk for AI-dependent engineering teams.
openads-ai
Advanced tools
Open-source AI command center for digital marketers. Audit campaigns, write ad copy, and build strategies — from your terminal.
██████╗ ██████╗ ███████╗███╗ ██╗ █████╗ ██████╗ ███████╗
██╔═══██╗██╔══██╗██╔════╝████╗ ██║██╔══██╗██╔══██╗██╔════╝
██║ ██║██████╔╝█████╗ ██╔██╗ ██║███████║██║ ██║███████╗
██║ ██║██╔═══╝ ██╔══╝ ██║╚██╗██║██╔══██║██║ ██║╚════██║
╚██████╔╝██║ ███████╗██║ ╚████║██║ ██║██████╔╝███████║
╚═════╝ ╚═╝ ╚══════╝╚═╝ ╚═══╝╚═╝ ╚═╝╚═════╝ ╚══════╝
AI Command Center for Marketers
Talk to your ad campaigns in plain English. Connect your Google Ads, Google Analytics (GA4), and Meta accounts, pick your favorite AI model, and let OpenAds handle the analysis while you focus on strategy.
OpenAds is an open-source CLI tool that turns any AI model into a marketing assistant. It's built for performance marketers, media buyers, and growth leads who want to audit campaigns, write ad copy, and build strategies — all from one place.
No code. No prompt engineering. No spreadsheet exports.
| Feature | What it means for you |
|---|---|
| 🧠 Pre-built marketing skills | The AI already knows Google Ads best practices, Meta creative formats, CRO frameworks, and copywriting rules. You just ask. |
| 🔌 Direct platform access | Connect your Google Ads, Google Analytics (GA4), and Meta accounts. The AI reads your live data — no more copy-pasting reports. |
| 🤖 Bring your own model | Use Google Gemini, OpenAI, Claude, or a local model running on your machine. Your choice. |
| 🛡️ Nothing goes live without you | The AI can read freely, but every write operation (campaign change, budget edit) requires your explicit approval. |
| ⚡ Three experience tiers | Express for local models, Standard for mid-tier, Full for frontier — the same beautiful UI adapts its depth to your model's capability. |
OpenAds automatically adapts its entire experience based on your AI model's capability. Same menus, same polish, fundamentally different depth.
┌────────────────────────────────────────────────────────────┐
│ ⚡ Express 📊 Standard 🚀 Full │
│ Llama 8B Gemini Flash GPT-4o │
│ Mistral 7B GPT-4o Mini Claude Sonnet │
│ Phi-3 Llama 70B Gemini Pro │
├────────────────────────────────────────────────────────────┤
│ System prompt: System prompt: System prompt: │
│ ~150 words ~300 words ~500 words │
│ │
│ Skills: compact Skills: full Skills: full │
│ (~100 words each) (~400 words each) + advanced │
│ │
│ MCP tools: none MCP tools: 6 MCP tools: 11 │
│ (context-based) (read-only) (read + write) │
│ │
│ Menu: 7 items Menu: 9 items Menu: 10 items │
│ AR: 4 commands AR: 8 commands AR: 12 commands │
│ │
│ Output: bullets Output: tables Output: deep │
│ & quick tables & structured analysis with │
│ reports cross-platform │
└────────────────────────────────────────────────────────────┘
A marketing manager running Llama 3.1 8B locally shouldn't get a worse experience — they should get a different but equally polished experience. Express mode delivers:
When you're ready for more depth, just run openads → Settings → Change experience tier.
Here is a look at OpenAds in action:
OpenAds is a local desktop application that runs in your computer's Terminal (a text-based window where you can run commands). Follow these simple steps to get started:
OpenAds runs on your computer using Node.js. If you don't have it yet, installing it is just like any normal application:
Cmd + Space (Spotlight search), type Terminal, and press Enter.Windows Key on your keyboard, type cmd (Command Prompt), and press Enter.Copy the command below, paste it into your Terminal window, and press Enter:
npm install -g openads-ai
💡 Permissions Error? If your Terminal shows a red error about "EACCES" or permissions, copy and paste this command instead:
sudo npm install -g openads-ai(Mac will ask you to type your computer password and press Enter).
Paste this command into your Terminal and press Enter:
openads setup
This launches a 6-step interactive setup wizard where you can:
Whenever you want to audit your campaigns or write copy, just open your Terminal, type:
openads
and press Enter to open your dashboard.
If you're already familiar with node packages, just run:
npm install -g openads-ai
openads setup
openads
# 1. Install Ollama (free, open-source)
brew install ollama # Mac
# or: curl -fsSL https://ollama.com/install.sh | sh # Linux
# 2. Pull a model
ollama pull llama3.1
# 3. Install OpenAds and pick "Local AI" + "Express" in setup
npm install -g openads-ai
openads setup
# 4. Run — no internet required
openads
Here are some real examples — just type what you need:
| You type | What happens |
|---|---|
Audit my Google Ads account and flag budget waste | Reads your live campaign data, finds underperforming keywords, and tells you where you're losing money. |
My Meta ROAS dropped 30% this week — what changed? | Pulls your Meta Ads data, compares to the prior period, and pinpoints what shifted. |
Write a 30-second video ad script for TikTok | Generates a hook → story → CTA script formatted for vertical video with platform-specific timing. |
| You type | What happens |
|---|---|
Write 5 Google Ads headlines for my product | Generates headlines under 30 characters using your product context, with multiple creative angles. |
Rewrite this landing page to be more persuasive | Applies PAS/AIDA frameworks, tightens the copy, and fixes benefit vs. feature balance. |
| You type | What happens |
|---|---|
Build a go-to-market plan for my Q3 launch | Produces a structured GTM playbook covering positioning, channels, budget, and timelines. |
Who are my top 3 competitors and what are they saying in their ads? | Analyzes competitor positioning, identifies messaging gaps, and recommends differentiation angles. |
Research my target audience for a B2B SaaS product | Builds a customer research brief: pain points, buying triggers, objections, and voice-of-customer language. |
| You type | What happens |
|---|---|
My landing page converts at 1.2% — how do I improve it? | Runs a CRO audit: checks message match, CTA placement, form length, and gives prioritized fixes. |
Set up an A/B test for my signup page headline | Designs a proper experiment with hypothesis, control vs. variant, sample size, and success criteria. |
Run autoresearch on my ad headlines overnight | The AI autonomously generates variants, scores them, keeps the best, and reports back in the morning. |
| You type | What happens |
|---|---|
Write a 5-email welcome sequence for new signups | Creates a full drip sequence: delivery → value → story → objection handling → soft pitch. |
Autoresearch is OpenAds' marquee feature: an autonomous loop that generates, scores, and iterates on marketing hypotheses without you lifting a finger.
The menu adapts to your tier:
| Tier | Available Commands | What you get |
|---|---|---|
| ⚡ Express | Generate, Plan, Debug, Fix | 4 commands, flat menu, 2 loop cycles |
| 📊 Standard | + Discover, Analyze, Ship | 8 commands, 5 phases, 3 loop cycles |
| 🚀 Full | + Validate (Predict, Probe, Reason, Scenario) | 12 commands, 6 phases, 5 loop cycles |
Full tier exclusive: The Validate phase includes multi-persona debates (5 expert personas evaluate your idea), stress-tests (8 personas attack your brief), and adversarial reasoning on key strategy calls.
OpenAds remembers what it learns about your business. After each conversation, the AI appends key insights to a plain markdown file at ~/.openads/context/my-business.md:
You can open and edit this file anytime — it's your data, not a black box. The longer you use OpenAds, the better its advice gets.
Set up automated campaign checks that run in the background — no server required.
openads schedule
| Preset | Frequency |
|---|---|
| 📊 Daily campaign health check | Every day at 8 AM |
| 💸 Budget pacing alert | Every 6 hours |
| 📉 Performance drop alert | Twice daily (9 AM & 5 PM) |
| 📋 Weekly performance report | Every Monday at 9 AM |
| ⏰ Custom (describe in plain English) | You choose |
Reports are saved to ~/.openads/reports/ in both Markdown and premium HTML dashboard formats. You can view, list, and open your reports directly:
openads report # List all generated reports
openads report [name] # Open a beautiful HTML dashboard in your browser
Manage your schedules:
openads schedule # Open the schedule manager
openads schedule list # See active schedules
openads schedule remove # Remove a schedule
Uses your OS scheduler (macOS launchd / Linux crontab) — works even when your terminal is closed.
To align with our Radical Simplicity & Safety by Default principles, OpenAds operates in two visual modes to give you total control and peace of mind over your ad accounts:
Configure your default mode anytime by running openads → Settings → Change operational mode.
Run the built-in diagnostics to check your setup:
openads doctor
This verifies your config file, API keys, platform connections (live token checks), required tools like uvx, and local AI readiness (checks if Ollama is running and your model is available).
~/.openads/ on your hard drive. They never touch our servers.Y.npm install -g openads-ai)--print mode for CI/CD pipelinesWe want OpenAds to be the standard open-source tool for AI-assisted marketing. You don't need to be a developer to contribute — marketing playbooks and strategy templates are just as valuable as code.
Read CONTRIBUTING.md to get started.
MIT.
Built on Pi (MIT). Includes tools derived from adloop (MIT) by kLOsk. Marketing skills inspired by marketingskills (MIT) by Corey Haines. Memory and background automation concepts inspired by Hermes Agent by Nous Research. Token optimization proxy concepts inspired by RTK by rtk-ai.
FAQs
Open-source AI command center for digital marketers. Audit campaigns, write ad copy, and build strategies — from your terminal.
We found that openads-ai demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 1 open source maintainer collaborating on the project.
Did you know?

Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.

Security News
The Fable shutdown shows how quickly model access can become a business continuity risk for AI-dependent engineering teams.

Security News
AI agents are pulling packages into environments no scanner is watching, creating exposure before security teams can see it.

Security News
GitHub Actions checkout now blocks risky pull_request_target checkouts by default to help prevent pwn request supply chain attacks.