Marketing · Copy, launches, pricing pages, ads, search. · a tool
The open-source project claude-ads, by AgriciDaniel
“Claude-first paid-media operations skill for Claude Code across 12 ad platforms (Google, Meta, YouTube, LinkedIn, TikTok, Microsoft, Apple, Amazon, Reddit, Pinterest, Snapchat, X): source-grounded…” — the project's own words, on GitHub
Paste this into the AI that runs on your computer. It does the install, checks it works, and tells you what to type first. If anything fails, paste the error back to it.
Install claude-ads (https://github.com/AgriciDaniel/claude-ads) on this computer and get it working for me. I am not a developer; explain each step in one line as you go and never skip one. You are Claude Code, running on my machine. 1. Check what this machine already has (git, Python or Node as the project needs, Docker if the README says so). Tell me anything missing and install it, asking me before anything that needs my password. 2. Clone https://github.com/AgriciDaniel/claude-ads into ~/tools/claude-ads and follow the README's install exactly. 3. Configure it. If it needs an API key or a login, stop and ask me for it; never guess one and never store it anywhere except where the README says. Tool-specific notes: Clone it with `git clone https://github.com/AgriciDaniel/claude-ads.git`, then run `bash install.sh --source=local` (or, inside Claude Code, `/plugin marketplace add AgriciDaniel/claude-ads` then `/plugin install claude-ads@ai-marketing-hub-claude-ads`). It needs Python 3.11 or 3.12: `python3.12 -m venv .venv`, then `.venv/bin/python -m pip install --no-deps -e .` and `.venv/bin/python -m pip install --require-hashes --only-binary=:all: -r requirements.lock`. Verify with `.venv/bin/python -m pytest -q` and `python -m claude_ads_core --version` — both should pass before you call it installed. Ask for each ad platform's credentials one at a time and put them in environment variables or the OS keychain, never in a file the repo tracks. PDF reports need WeasyPrint and Pango on the machine — skip that step if Markdown or HTML output is enough. 4. Run the smallest test the README gives, and show me the output. 5. When it works, tell me what to type first, in one line, for this job: Run an account audit. If anything fails, show me the exact error and fix it before going on.
Install claude-ads (https://github.com/AgriciDaniel/claude-ads) on this computer and get it working for me. I am not a developer; explain each step in one line as you go and never skip one. You are Kimi, running on my machine. 1. Check what this machine already has (git, Python or Node as the project needs, Docker if the README says so). Tell me anything missing and install it, asking me before anything that needs my password. 2. Clone https://github.com/AgriciDaniel/claude-ads into ~/tools/claude-ads and follow the README's install exactly. 3. Configure it. If it needs an API key or a login, stop and ask me for it; never guess one and never store it anywhere except where the README says. Tool-specific notes: Clone it with `git clone https://github.com/AgriciDaniel/claude-ads.git`, then run `bash install.sh --source=local` (or, inside Claude Code, `/plugin marketplace add AgriciDaniel/claude-ads` then `/plugin install claude-ads@ai-marketing-hub-claude-ads`). It needs Python 3.11 or 3.12: `python3.12 -m venv .venv`, then `.venv/bin/python -m pip install --no-deps -e .` and `.venv/bin/python -m pip install --require-hashes --only-binary=:all: -r requirements.lock`. Verify with `.venv/bin/python -m pytest -q` and `python -m claude_ads_core --version` — both should pass before you call it installed. Ask for each ad platform's credentials one at a time and put them in environment variables or the OS keychain, never in a file the repo tracks. PDF reports need WeasyPrint and Pango on the machine — skip that step if Markdown or HTML output is enough. 4. Run the smallest test the README gives, and show me the output. 5. When it works, tell me what to type first, in one line, for this job: Run an account audit. If anything fails, show me the exact error and fix it before going on.
Install claude-ads (https://github.com/AgriciDaniel/claude-ads) on this computer and get it working for me. I am not a developer; explain each step in one line as you go and never skip one. You are Gemini CLI, running on my machine. 1. Check what this machine already has (git, Python or Node as the project needs, Docker if the README says so). Tell me anything missing and install it, asking me before anything that needs my password. 2. Clone https://github.com/AgriciDaniel/claude-ads into ~/tools/claude-ads and follow the README's install exactly. 3. Configure it. If it needs an API key or a login, stop and ask me for it; never guess one and never store it anywhere except where the README says. Tool-specific notes: Clone it with `git clone https://github.com/AgriciDaniel/claude-ads.git`, then run `bash install.sh --source=local` (or, inside Claude Code, `/plugin marketplace add AgriciDaniel/claude-ads` then `/plugin install claude-ads@ai-marketing-hub-claude-ads`). It needs Python 3.11 or 3.12: `python3.12 -m venv .venv`, then `.venv/bin/python -m pip install --no-deps -e .` and `.venv/bin/python -m pip install --require-hashes --only-binary=:all: -r requirements.lock`. Verify with `.venv/bin/python -m pytest -q` and `python -m claude_ads_core --version` — both should pass before you call it installed. Ask for each ad platform's credentials one at a time and put them in environment variables or the OS keychain, never in a file the repo tracks. PDF reports need WeasyPrint and Pango on the machine — skip that step if Markdown or HTML output is enough. 4. Run the smallest test the README gives, and show me the output. 5. When it works, tell me what to type first, in one line, for this job: Run an account audit. If anything fails, show me the exact error and fix it before going on.
Install claude-ads (https://github.com/AgriciDaniel/claude-ads) on this computer and get it working for me. I am not a developer; explain each step in one line as you go and never skip one. You are Codex, running on my machine. 1. Check what this machine already has (git, Python or Node as the project needs, Docker if the README says so). Tell me anything missing and install it, asking me before anything that needs my password. 2. Clone https://github.com/AgriciDaniel/claude-ads into ~/tools/claude-ads and follow the README's install exactly. 3. Configure it. If it needs an API key or a login, stop and ask me for it; never guess one and never store it anywhere except where the README says. Tool-specific notes: Clone it with `git clone https://github.com/AgriciDaniel/claude-ads.git`, then run `bash install.sh --source=local` (or, inside Claude Code, `/plugin marketplace add AgriciDaniel/claude-ads` then `/plugin install claude-ads@ai-marketing-hub-claude-ads`). It needs Python 3.11 or 3.12: `python3.12 -m venv .venv`, then `.venv/bin/python -m pip install --no-deps -e .` and `.venv/bin/python -m pip install --require-hashes --only-binary=:all: -r requirements.lock`. Verify with `.venv/bin/python -m pytest -q` and `python -m claude_ads_core --version` — both should pass before you call it installed. Ask for each ad platform's credentials one at a time and put them in environment variables or the OS keychain, never in a file the repo tracks. PDF reports need WeasyPrint and Pango on the machine — skip that step if Markdown or HTML output is enough. 4. Run the smallest test the README gives, and show me the output. 5. When it works, tell me what to type first, in one line, for this job: Run an account audit. If anything fails, show me the exact error and fix it before going on.
Point it at one ad account export and run the audit command — good output is a dated finding with a confidence level attached to each claim, not just a list of stats.
Open two or three of the findings it cites and confirm the underlying numbers match your own ads dashboard before you act on any of them.
Run the planning command for one channel and read the draft yourself — it should propose budget and channel changes, not make them, until you say so.
AgriciDaniel/claude-ads on GitHub, under the MIT licence. 9,072 stars, checked 9 September 2026. Last change 13 July 2026. We did not write it; we checked it, and wrote this page so you can use it.
gstack turns Claude Code into a small engineering team with defined roles — a CEO for direction, an engineer, a designer, a QA lead, a security reviewer and a release engineer — each one a slash command with its own review gate.
marketingskills is a set of reference playbooks your AI coding agent reads before doing marketing work — conversion optimization, copywriting, SEO, analytics and growth engineering — so instead of guessing at best practice, it works from the same frameworks a marketing consultant would use.
toprank connects your AI agent to your live SEO, Google Ads, Meta Ads and analytics accounts and gives it forty-five ready-made workflows for auditing and reporting on them — why organic traffic dropped, how a Google Ads account is actually performing, what an AI search engine says about you.
If you're launching or growing a brand across Asia and the West, a call is where we work out whether it's a fit.