Price Competitor Scan
by
Competitive price intelligence for ecommerce teams: track rival pricing, detect the changes that actually move your margin, and get a positioning call you can act on the same day.
Manually tracking competitor pricing across a dozen browser tabs is slow, and it breaks the moment someone goes on holiday. Worse, the shifts that actually matter (a quiet promo code, a switch from MSRP to a checkout-only discount, a size going out of stock) are the ones a human eye misses. This skill turns that into a repeatable scan: it reads a competitor product page, extracts unit price, currency and tax status, picks up active discounts and visible promo codes, records stock and backorder signals, and writes the whole thing to a local snapshot. On the next run it diffs against the last snapshot and tells you what changed, by how much, and what to do about it.
What It Does
- Extracts unit price, currency and tax status from the product page DOM or raw text.
- Identifies active discounts, sale prices and visible promo codes.
- Reads inventory signals: in stock, low stock, backorder, sold out.
- Writes a dated snapshot to local JSONL so history is yours, not a vendor's.
- Generates a delta report with absolute and percentage change since the last scan.
- Recommends a positioning move: match, hold, or let them clear stock.
Use Cases
- Watch rival product pages for clearance sales and 10%+ price drops.
- Ship a weekly competitive memo to the pricing committee without touching a spreadsheet.
- Spot free-shipping threshold gaps where a competitor offers better logistics value.
- Time an aggressive promo against a competitor stock-out window.
Example Session
Compare item-123 at example.com to our $49.99, check for promos, and diff it against last week.
Three of the five pages you watch moved since Thursday. Here is the diff against your last snapshot.
- Product | Our price | Competitor | Delta | Promo | Stock
- Item-123 | $49.99 | $44.00 | -12% | FALL20 | In stock
- Item-450 | $29.00 | $29.00 | 0% | None | Low stock
- Item-771 | $115.00 | $129.00 | +12% | None | Sold out
- Item-902 | $76.50 | $74.90 | -2% | None | In stock
Item-123 lost 12% while stock stayed high: that reads as clearance, not repositioning. Hold your price and re-scan in seven days. Item-771 is sold out at a 12% premium, which is the cleanest promo window you have this month.
Why This Beats Prompting It Yourself
Raw Answers vs Sourced Numbers: The Hidden Risk
A generic prompt returns a price with nothing behind it. Ask whether a competitor is cheaper and you get “around $45, roughly 10% below you”, with no timestamp, no stock, and no distinction between the MSRP and what is actually charged at checkout. The model picks one number and never tells you which. Nothing in the prompt stops it, because a prompt is about phrasing, not evidence.
When You Still Need a Snapshot Workflow
The moment a price decision crosses two dates, you need one place where the reading, its timestamp and its raw source live together. This skill writes $44.00 read at 09:12, against $50.00 on Sept 11, with FALL20 active and stock still high, so the call is clearance rather than repositioning. Every figure traces back to the page text it came from, which is what makes the delta defensible in a pricing meeting.
Known Limitations
It needs product URLs that don't block headless browsing or basic scraping. Heavily bot-protected sites will fail loudly rather than guess. Snapshots must be stored locally as JSONL for historical deltas to work, so the first run has nothing to compare against.
Install It
- Download the ZIP.
It is free and there is no account to create. - Unzip it into your agent's skills folder.
Claude Code reads ~/.claude/skills/, which is hidden by default: the command in the folder block opens it. Other agents scan their own directory, so drop the same folder there instead. - Ask your agent to use it.
Restart the agent if it was already running, then it picks the skill up with no config.
skills folder
Copy
~/.claude/skills/price-competitor-scan/
SKILL.md
references/selectors.md
snapshots/ # JSONL, one line per scan
# macOS: create the folder and open it in Finder
mkdir -p ~/.claude/skills && open ~/.claude/skills
# Windows: paste in the Explorer address bar
%USERPROFILE%\.claude\skills\Frequently asked questions
How is this different from a standard web scraper?
A scraper hands you a page. This hands you a decision. It normalizes unit price, currency and tax status, separates the MSRP from the price actually charged at checkout, diffs the result against your own history, and ends on a positioning call: match, hold, or let them clear their stock.
Which agents does the skill run in?
Any agent that supports the open SKILL.md format: Claude Code, Cursor, Codex CLI, GitHub Copilot, Gemini CLI, Manus, Grok and others load it unmodified. The format is portable, the location is not. Each agent scans its own skills directory, so you drop the same folder into whichever one yours uses. There is no config file to edit and no API key to provision.
What setup is needed to track price changes over time?
One folder. Every scan appends a line to a local JSONL snapshot next to the skill, so the second run already has a baseline to compare against. Nothing is uploaded, and the history stays yours rather than a vendor's.
What does a scan actually capture?
Unit price, currency and tax status, sale price, any promo code visible on the page, stock and backorder state, shipping thresholds, and for each of those the raw text it was extracted from plus a timestamp. The evidence travels with the number, which is what makes the delta defensible in a pricing meeting.
Are there sites it cannot monitor?
Yes, and that is deliberate. Pages behind aggressive bot protection or a login wall make the scan fail loudly instead of returning a number the model guessed. A missing data point is recoverable; a confident wrong price that reaches your pricing committee is not.
What is included with the skill?
The skill itself, the selector reference for the marketplaces ecommerce teams watch most, the snapshot schema, and the report template. It is free and security scanned. We re-publish the ZIP when the marketplaces change their markup, so download it again if a scan starts missing fields.
A skill is only as good as the data underneath it.
Polar reconciles 45+ sources in one semantic layer, so your agent answers on the same numbers your team reports on.
<script type="application/ld+json">{"@context":"https://schema.org","@type":"FAQPage","mainEntity":[{"@type":"Question","name":"How is this different from a standard web scraper?","acceptedAnswer":{"@type":"Answer","text":"A scraper hands you a page. This hands you a decision. It normalizes unit price, currency and tax status, separates the MSRP from the price actually charged at checkout, diffs the result against your own history, and ends on a positioning call: match, hold, or let them clear their stock."}},{"@type":"Question","name":"Which agents does the skill run in?","acceptedAnswer":{"@type":"Answer","text":"Any agent that supports the open SKILL.md format: Claude Code, Cursor, Codex CLI, GitHub Copilot, Gemini CLI, Manus, Grok and others load it unmodified. The format is portable, the location is not. Each agent scans its own skills directory, so you drop the same folder into whichever one yours uses. There is no config file to edit and no API key to provision."}},{"@type":"Question","name":"What setup is needed to track price changes over time?","acceptedAnswer":{"@type":"Answer","text":"One folder. Every scan appends a line to a local JSONL snapshot next to the skill, so the second run already has a baseline to compare against. Nothing is uploaded, and the history stays yours rather than a vendor's."}},{"@type":"Question","name":"What does a scan actually capture?","acceptedAnswer":{"@type":"Answer","text":"Unit price, currency and tax status, sale price, any promo code visible on the page, stock and backorder state, shipping thresholds, and for each of those the raw text it was extracted from plus a timestamp. The evidence travels with the number, which is what makes the delta defensible in a pricing meeting."}},{"@type":"Question","name":"Are there sites it cannot monitor?","acceptedAnswer":{"@type":"Answer","text":"Yes, and that is deliberate. Pages behind aggressive bot protection or a login wall make the scan fail loudly instead of returning a number the model guessed. A missing data point is recoverable; a confident wrong price that reaches your pricing committee is not."}},{"@type":"Question","name":"What is included with the skill?","acceptedAnswer":{"@type":"Answer","text":"The skill itself, the selector reference for the marketplaces ecommerce teams watch most, the snapshot schema, and the report template. It is free and security scanned. We re-publish the ZIP when the marketplaces change their markup, so download it again if a scan starts missing fields."}}]}</script>