Hire an AI product candidate test planner

Everything it can hand back

Your task sets the actual outputs. The specialist picks the methods that cover them.

• One pick with its score out of 100, its rival check, its category call and why it beat the rest, or a clear "no pick this month"
• A test plan with a spending cap, a pass result and a kill rule
• A watch list showing what would move each product
• A list of dropped products with one reason each
• What to bring next month

How it does the job well

• Start from the user's task and the content, then the layout. Decoration comes last.
• Design every state: empty, loading, error, long text and mobile.
• Keep one visual system. Reuse spacing, type and colour tokens instead of one-off values.
• Check text contrast (at least 4.5:1 for body text) and keyboard focus as you go.

Mistakes it avoids

• Calling a product a winner from a score alone. A score says how open and how profitable it looks, not that it will sell.
• Picking a product just because the month needs a pick.
• Re-scoring the same candidates every month with no new evidence.
• Running the deep checks on everything. Only the few that pass earlier gates get them.
• Designing only the happy path at desktop width.
• Copying a reference's logo, text or claims instead of learning from its structure.

What a good result looks like

An example, not your task: You want a steady way to find your next product to sell, not a one-off hunch.

• This month's pick. One product with its score, field call, category call, launch call and why it beat the rest, or no pick.
• Test plan. The cheapest test, the spending cap, the pass result and the kill rule.
• Watch list. Products kept on watch, with what would move each one.
• Dropped this month. Each dropped product with a one-line reason.
• Needs from you. Costs or evidence still missing so next month can score them.
• scout/pipeline.md. One row per candidate with score, field verdict, status and the reason for the last change.

How the work is checked

• Every score came from the scoring skill, with the sums shown. You did not score by feel.
• No candidate was scored without a landed cost and a target price.
• No more than one pick, and no pick when new-product-line-checker said NO-GO.
• Every drop and every watch has a reason in the pipeline.
• States which checks passed, failed or could not run, and names missing inputs.

These checks have not been run against a model yet. Your AI uses them to check its own work.

Capabilities

• Read the pipeline. Opens last month's list of candidates and finds the ones on watch.
• Re-check the old ones. Checks each watched product for new rivals. A product that fell to skip is dropped with the reason.
• Find and score new ones. Reads your evidence for proven products and weak-rated gaps, then scores each new candidate that has a cost and a price.
• Check the field. Candidates that pass the score get a rival check. Crowded ones only stay if a buyer group nobody targets turns up.
• Check the category. For top candidates in a category you do not sell in yet, checks market size and trend risk.
• Plan one test. Picks the single best one and plans a capped test with a kill rule. If it fails the check, there is no pick this month.

How the role works

1. Checks your task, context and available tools.
2. Picks the methods needed and prepares a work plan.
3. Carries out supported tasks, with each method's limits and review gates.
4. Checks the real deliverables and corrects supported faults.
5. Hands back files, evidence and unresolved gaps.

Your specialist's working setup

Download the base specialist without a task form or account. README.md and SETUP.md explain how to use the folder. SOUL.md defines its scope. INTAKE.md asks missing questions inside your AI. PLAYBOOK.md, SKILLS.md, BEST_PRACTICES.md, CHECKS.md and HANDOFF.md guide the work and its checked deliverables.

Complete associated skills, supporting files and licences are included. Choose only the methods needed for the actual task. Keep context, working files and accepted feedback in your AI project.

Use this specialist through MCP or the API

Call prepare_specialist with role_id monthly-product-opportunity-scout only. No task is required to load it. Your AI asks what job you need done, what you want back and which relevant context it can use.

For task-led hiring, call find_specialist, explain the suggested specialists and let the user accept or change the team.

Complete ZIP: Download base specialist. API: GET /api/v1/specialists/monthly-product-opportunity-scout/pack. The host AI supplies its own model, tools and execution; a pack does not connect accounts or run a task.

Inputs and helpful context

The specialist starts from the task brief. Only active methods require their inputs.

For the primary method, when used

• Your market and who buys, in a short paragraph
• Screenshots from the Meta Ad Library, sales rankings or product graphs for your market (the first month, then fresh ones each month). The Meta Ad Library is free to search. Sales-ranking and product-graph tools may need a paid account, or you can save screenshots by hand. The agent pays for and scrapes nothing
• Your expected landed cost per unit (product plus shipping to the customer), your target price and the most you can lose on a test
• Last month's scout/pipeline.md, if you ran it before (a pasted copy works in a chat)
• Customer reviews of the category (optional)
• A list of what you already sell (optional)

Helpful context (optional)

• Screenshots of product graphs or sales rankings for your market
• Results of any earlier product tests you ran

Tools and access

• Claude Code: It can keep the scout folder and pipeline, read last month's files and run every step on a schedule.
• Meta Ad Library: Capture the ads for each candidate here, so the rival check has real counts.
• Meta Ad Library: Free. It shows active ads, start dates and image or video formats for any advertiser.
• A paid ad spy tool: Needed for the filters on days running, active ads per brand and impressions. Any tool with those filters works.
• Google Lens: Free. It runs the reverse image search that finds the supplier listing for a product photo.
• Any chat AI with web search: Needed to find reviews, forum posts and Q&A. Without it, paste the text in.
• Amazon reviews and Q&A: Free text from real buyers on what products get wrong.
• Google Trends or a keyword tool: Gives search volume to confirm a gap.
• Any good chat AI: Any current chat AI can run the sums and the rule marks. Paste your costs and links.
• Meta Ad Library: Free. Use it to find videos or ads for the content rule and to see if rivals run the product.
• Google Trends: Free. Check the search line for the exact product to see if it is a trend or evergreen.
• Meta Ad Library: Free. It lists active ads from any advertiser, which is what the 7-day store count needs.
• Any good chat AI: Any current chat AI can count what you paste in and apply the limits.
• Meta Ad Library: Free. Shows which rivals run ads and who those ads speak to.
• TikTok search: Shows the clips rivals post and who appears in them.
• Any good chat AI: Any current chat AI can run the steps if you paste the rival ads in.
• Google Trends: Free. Shows a 24-hour and a multi-year view, and compares terms.
• Amazon best-seller lists: Free. Check ranks daily to see what is rising now.
• Any good chat AI: Any current chat AI can run the checks if you paste the data in.
• Shopify reports: Show which products your existing buyers bought together, for the cross-sell check. You need a paid Shopify plan.
• Klaviyo or another email tool: Runs the first test, an email send to your list. Most email tools are paid.
• Any good chat AI: Any current chat AI can run the checks from the facts you paste.

Functions and deliverables

• Product candidate test planner: Review product candidates from collected evidence. Check costs and competition, then propose one small test or no pick.
  • One pick with its score out of 100, its rival check, its category call and why it beat the rest, or a clear "no pick this month"
  • A test plan with a spending cap, a pass result and a kill rule
  • A watch list showing what would move each product
  • A list of dropped products with one reason each
  • What to bring next month
• Advertised product finder: Find proven products to sell. Name your market and get ad spy filters, an ad reading routine and a short list of five to ten products marked winner or fluke.
  • Filter settings you can copy into your tool, or the free searches to run
  • A list of 5 to 10 candidates with advertiser, ad age, ad count, price and angle
  • The longest-running ad to study for each, and what to change so you are not a clone
  • A winner or fluke call for each, a supplier cost check and the rivals to look at
• Product opportunity reviewer: Find product gaps where buyers search a lot but rate what exists low. Give it a category and get ranked pain points with proof links.
  • The top ten unmet pain points, grouped by theme, ranked by urgency and willingness to pay, with proof links
  • Sub-categories with a low average rating but strong search
  • Rank, review and seller data where you supplied it or the AI could find it
  • A product concept for one narrow buyer group, as a mood board and not a design
• Product idea viability scorer: Score a winning product idea before you buy stock or run ads. Give it your cost and price. You get 11 rule marks, a score out of 100 and a verdict.
  • The margin sums, with the multiple and the gross profit per unit
  • Each of the 11 rules marked pass, warn or fail, with the points
  • A score out of 100 with the sum shown, and a verdict of go, test small or drop
  • The one thing to fix first, and a one-line record for your shortlist
• Competitor saturation checker: Check how crowded a product is before you test it. Give it rival ads from the Ad Library. You get a score out of 100 and a go, test fast or skip.
  • A count of stores advertising it, how many are serious, and how many started in the last 7 days
  • A niche-level crowding note, not only the exact product
  • A score out of 100 with the sum shown, and a go, test fast or skip
  • The rival pages to watch, and their live ads if you pasted them
• Untapped buyer finder: Find buyer groups your rivals miss for a crowded product. Give it rival ads. You get ranked groups and a small test for each.
  • A map of who rivals target and who they miss
  • Three to five untapped buyer groups, each with the problem it has
  • One test plan per group: a hook, a creative idea and a landing page line
  • A note on which groups to test first for your budget
• Category and trend risk checker: Name a category to enter. Get a score out of 100 on market size, trend risk and fit.
  • A score out of 100 with the weak spots named
  • A trend note: fad, steady or growing, and a seasonality note
  • Demand size and where buyers are
  • Go, hold or drop, and a way to widen your range if you need one
• New product line checker: Share a new product, margin and launch budget. Assess its role, buyer overlap and sales channels, then draft a capped test.
  • A label: CAC product or LTV product
  • A score out of 100 on the three checks, with the sum shown
  • Go or no-go, with the reason
  • A capped test plan with kill rules
  • The condition a buyer must meet before you place the minimum order
• Brand profile builder: Share your website. Get a reusable profile covering your products, customers, voice and approved claims, with sources.
  • A folder of eight plain files, from a start file for your AI to a claims list and a change log
  • A tag on every fact, and every claim marked approved, pending or banned
  • A two-sentence diagnosis of your biggest gap, and a list of what's missing and who can answer it
  • A short slice for your next step, if you name one

Tasks you can use it for

• You want a steady way to find your next product to sell, not a one-off hunch.
• You have a list of product ideas and want to know which are worth a small test.
• A product on your watch list may have filled up with rivals and you want to know before you order.
• You saved ad library captures or sales rankings and want them turned into a pick and a plan.

Limits to check before starting

• You have already chosen the product. Use the New product line checker to judge it and plan the test.
• You have one idea and want a quick yes or no on price and margin. Use the Product idea viability scorer.
• You have no captures and no cost figures. It cannot browse, so it has nothing to read. Collect the ads and rankings first.
• You sell on Amazon. Use the Amazon product viability scorer for that marketplace.

Best way to use this role

Give it one clear outcome, the actual files and a definition of done. Keep the work in one AI project so it can reuse checked context and earlier files.

• Add the agent to your AI. In Claude Code, save it in .claude/agents or add it as a skill. In Claude, ChatGPT or Gemini, paste it at the start of a chat. It saves a scout folder, so Claude Code is easiest; in a plain chat, paste last month's pipeline and save the new one yourself.
• Save your evidence first. Capture ads for each product from the Meta Ad Library, and screenshot a sales ranking or product graph. The agent cannot browse the web.
• Type your market, buyers, landed cost, target price and the most you can lose on one test. Mark any cost you are unsure of as a guess.
• Read the pick. Check that every score shows its sums, and that each dropped or watched product has a reason.
• Decide on the test yourself. The agent never orders, lists a product or spends money. When you run the test, add the result to the pipeline.
• Next month, run it again with fresh captures and last month's pipeline. Take the pick to the New product line checker for a full launch check.

Read it as data

Read the complete role setup as JSON. Connect your AI to the AI Blocks MCP library and read aiblocks://roles/monthly-product-opportunity-scout for its role instructions and pinned sources.

This is a prepared AI role. No task has run. Your AI carries out authorized work using its available tools and reports the actual deliverables and checks.