Prompt library for the AMALYTIX AI Agent
The colleague you've been missing on your Amazon account. Prompts to copy, adapt, and save.
Version 2 of the AMALYTIX AI Agent is here. Same chat as before, now backed by more of your Amazon data and Amazon knowledge. Ask the questions no report could ever cover. Then let the agent do the work: fix a listing, send the update live, automate the tasks you repeat.
The prompts below are examples that get real results, not the full map. Build your own for your use case and save them in AMALYTIX under Settings > Agents.
Replace placeholders such as <ASIN> or <period> before you send.
38 prompts
Daily routine
Create my daily briefing.
What are the most critical problems in my account right now?
What happened in the last 24 hours?
Show me all high-priority alerts from the last 14 days.
Sales and diagnosis
Analyze my sales data for <period> across all marketplaces and name the three biggest movers up and down.
Why did <ASIN> lose revenue in <period> versus <comparison period>? Look at glance views, Buy Box, conversion rate, price and competition.
Which marketplaces and which products caused the change this month?
Compare week <N> with week <N-1> for <marketplace> and tell me the winners and losers.
Analyze the return rates per marketplace since <date>. Which ASINs drive them in units and value?
Listing content
<ASIN>: do a full listing optimization. Suggest a 75 character title and a 125 character item highlight, and review bullets, description and backend keywords. Do not transfer until I approve.
Changes data, after your approval
Which keywords bring traffic to this ASIN but are missing from the listing?
Shorten the title to 75 characters and move the removed keywords into the item highlight.
Changes data, after your approval
Clean the backend keywords: remove terms already in the title, bullets or description, remove internal duplicates, and fill up to the byte limit.
Changes data, after your approval
Use <ASIN> as the reference. Apply its content to all child ASINs and change only the product-specific attributes.
Changes data, after your approval
Translate this snapshot (title, bullets, description, keywords) into all marketplaces where the product is listed, respecting the byte limits.
Changes data, after your approval
Here is the manufacturer product information: <paste>. Create SEO-optimized Amazon content from it.
Changes data, after your approval
Catalog audits
Which of my high-revenue products have short or missing titles?
Check my top 50 ASINs in <marketplace> for listing errors and give me a table of ASIN, title and error.
Which products have fewer than five bullet points?
Show me all ASINs where the attribute <name> is missing or wrong.
Which articles are listed in <market A> but not in <market B>? Match on EAN.
Keywords and competition
What are my biggest keyword ranking losers this week, and what should I do about them?
Give me the Amazon search volume for these search terms in <marketplace>: <list>.
Analyze the competitor content for this product and tell me what they cover that I do not.
Which competitor wins the most shared keywords with me?
Advertising
Which campaigns in <marketplace> have high spend and low ROAS in the last 30 days?
Which search terms spent budget in the last 30 days without a single order?
Which of my high-revenue products are not advertised at all?
Are there duplicate keywords running in several campaigns?
I have <budget> per month for all campaigns. How should I split it across portfolios, and why?
Inventory and Buy Box
Which products will stock out in the next 14 days?
How many days of stock do my top sellers have?
How high is my risk of long-term storage fees?
Where did I lose the Buy Box in the last seven days, and to whom?
Working with the Agent
Save this analysis as a document in folder <name>.
Give me the table as a document I can download.
Summarize the current state as a prompt I can paste into a new conversation.
Do not change anything yet. First audit <scope> and report back, then wait for my approval.
Try it yourself
What you can ask goes well beyond this list. Build around your own use case and see what the agent takes off your plate.