
AI Marketing Operations & Campaign Workflow Agent
Nastasya




AI agentAI Marketing Agents
Finds where your funnel leaks, audits the page, ranks the tests and sizes the experiment properly. Works with Claude &…
$14.99· one file, yours to keep
Yours to keep · instant download
Or all of them, $6.99/mo
Ambrogio and 2,500+ other specialists, in your AI. Ask for any of them by name.
7 days free, then $6.99/mo · cancel anytime
One file
All-in-one
30-day money-back · secure checkout by Stripe
# Ambrogio - AI Conversion Rate Optimization Agent ## What This Agent Does Ambrogio takes a funnel that is losing people and works out where, why, and what to do about it in what order. He does three things. He diagnoses: given the numbers at each step, he finds the stage with the largest recoverable loss rather than the lowest rate. He audits: given the page content, copy and structure the user pastes in, he lists the specific friction points and what each one costs. He designs: he turns those into hypotheses, ranks them, and specifies the experiment
Checkout funnel: 100k monthly sessions, 2.1% add to cart, 38% checkout completion. Which number do I attack first?
Rank by recoverable revenue, not by ugliest percentage - your 2.1% add-to-cart looks worse, but the 62% checkout abandonment sits on buyers with intent, so each recovered point there is worth multiples. We quantify both steps by conversions and value actually recoverable, then audit the checkout you paste in for clarity, trust, effort and mobile friction with the mechanism named for each issue. Obvious fixes ship without testing; real hypotheses get written with a believed cause, not a colour change. One correction: at 38% completion the culprit is almost certainly surprise costs or forced accounts. Is guest checkout available on mobile?
A CRO working document: a funnel diagnosis showing where the recoverable value sits, a friction audit of the page or flow you paste in grouped by clarity, trust, effort, motivation and technical issues, a ship-without-testing list, a ranked test backlog where each hypothesis names a cause, a full experiment design for the top test with sample size and duration, the tracking and customer research you are missing, and every assumption he had to make.
Paste the .md file into your Claude Project Instructions (or ChatGPT's system prompt or custom instructions), then paste the activation prompt into the conversation. Ambrogio starts asking questions immediately. Takes under two minutes.
Claude, ChatGPT, Gemini, Copilot, Perplexity, or any AI chat that accepts a system prompt or custom instructions. Ambrogio is a plain text agent file, not an app or a plugin, so there is nothing to connect and no API key required.
Ambrogio calculates that before recommending anything. If your traffic cannot detect the effect in a reasonable window he says so plainly, tells you the traffic you would need, and switches to the approach that fits: ship the high-confidence fixes on judgement with a before-and-after observation window, and use customer research and session review instead of an underpowered test that only manufactures false wins.
Download your .skill file from the order email or your account. In Claude, turn on code execution in Settings › Capabilities, then go to Customize › Skills › + › Create skill › Upload a skill. Start a new chat and ask for the specialist by name.