Scrape product catalogs from any e-shop and match the same products across shops, without EAN or GTIN. Hosted remote MCP with OAuth. Pay per result from prepaid credit.
Scored across 15 tools
Each tool maps to a distinct resource+action, and the descriptions explicitly say which tool to use at which stage of the pipeline. The only mild overlaps are within the file-retrieval family (scout_get_export vs scout_get_sample vs scout_get_url_results) and scout_list_exports vs scout_get_export, but the descriptions clarify the list-then-fetch relationship.
All 15 tools use the same scout_ prefix followed by a consistent verb_noun pattern (list_projects, get_job, create_project, approve_job, scrape_website, match_websites). No mixed casing or alternate verb styles.
15 tools is at the top of the well-scoped range but each one covers a necessary step of a genuinely multi-stage pipeline (projects, websites, jobs, approvals, samples, exports, matching). Nothing appears redundant or trivially thin.
The lifecycle is well covered: project creation, website onboarding, job creation/approval/cancellation/polling, sample and export retrieval, and balance checks. Gaps are around destructive/administrative operations — there is no delete or update for projects or websites, and no way to remove a website from a project.
Add a website to a project: analyze it and scrape a free sample of its products. Each customer has free analyses each month. After them, an analysis charges the prepaid balance when the job is created, without an approval. scout_get_balance shows the free analyses left and the price of one. If none are left, confirm the price with the user before you call this tool. Returns the setup job; poll it with scout_get_job(job_kind="setup") until it is COMPLETED, then check the sample with scout_get_sample. A job that FAILED with the error_code CURRENCY_NOT_AVAILABLE found that the website does not offer the currency: no switch of the website shows its prices in it. The error of the job names the currency the website shows, when the analysis found it. The analysis is used, also a paid one. Ask the user for a currency the website offers, then analyze the website again. A project without localization gets the country and currency of this website.
country: ISO 3166 code of the market to scrape, e.g. "CZ". currency: ISO 4217 code of the prices to scrape, e.g. "CZK".
Inputs: url (required), country (required), currency (required), project_id (required). Returns: result
Approve a job that waits for approval and start it. This charges the prepaid balance of the customer: the estimate now, and at most the cap of the estimate after the job. Confirm the price with the user before you call this tool. Fails with 402 when the spendable balance or the monthly spend limit is too low.
Inputs: job_id (required), job_kind (required). Returns: result
Cancel a job that waits for approval, e.g. when the user does not accept its price. Nothing is charged. A matching job that waits for approval also blocks the free sample. Fails with 409 when the job cannot be cancelled anymore, e.g. after its approval.
Inputs: job_id (required). Returns: result
Create a project only when the user asks for a new one, e.g. for another client or market. A project is a separate workspace with its own websites and exports. A matching connects only the websites of one project. Each project pays for its websites: a website in a second project takes another analysis and another paid scrape. Thus do not create a project for each website. The customer_id is in scout_list_projects.
country: ISO 3166 code of the market of the project, e.g. "CZ". currency: ISO 4217 code of the prices in the exports, e.g. "CZK".
Inputs: name (required), country (required), currency (required), customer_id (required). Returns: result
Get the prepaid balance of a customer: the total, the part held by running jobs, and the spendable rest. customer.quotas gives the free analyses left this month and the price of an analysis after them. The customer_id is in scout_list_projects.
Inputs: customer_id (required). Returns: result
Get a temporary URL of an export file of a project. Give the dataset of the file from scout_list_exports, and its website_id if the file has one. The file exists after the job is COMPLETED.
Inputs: dataset (required), project_id (required), website_id. Returns: result
Get the state of a job: status, estimate, approval, settlement, and error. Poll every 5 to 10 minutes until the status is COMPLETED, FAILED, or CANCELLED. The time depends on the number of products, because DeepScout fetches the pages at a rate that the e-shop accepts. The usual times, with the wait in the queue, are:
setup: from a few minutes to 3 hours.
scraping: 1 to 2 hours for a few thousand products, several hours for 20,000, and up to 1 day for 100,000 or more.
matching: 1 hour for 1,000 products, 4 to 15 hours for 10,000, 1 to 2 days for 50,000.
EXPORTING, the last status: up to 1 hour. scrape_progress.percent and matching_progress.percent show the progress. matching_progress.estimated_minutes_left shows the time left. In the queue and in some phases, the progress does not change. A job without an error is not stuck: tell the user about a problem only if the job FAILED.
Inputs: job_id (required), job_kind (required). Returns: result
Get a temporary URL of the sample products file of a website. The sample exists after the setup job of the website is COMPLETED.
Inputs: project_id (required), website_id (required). Returns: result
Get a temporary URL of the CSV file with the result of each URL of a scraping job of given URLs. A row has the url, its result, and the product_url of its product. The file exists after the job is COMPLETED.
Inputs: job_id (required). Returns: result
List the export files of a project: the scraped products and prices, and the matches. A download_url needs the API key. Get a temporary URL of a file with scout_get_export.
Inputs: project_id (required). Returns: result
List the 20 newest jobs of a kind in a project, e.g. to find the id of a matching job. website_id limits the setup and scraping jobs to one website.
Inputs: job_kind (required), project_id (required), website_id. Returns: result
List the projects of the user: id, name, localization, and the customer that pays for them. The exports of a project have prices in its localization currency. Most tools need a project_id.
Inputs: no named parameters. Returns: result
List the websites of a project, each with the job that defines its state. full_scrape_estimate is the price of a full scrape now. The API calculates it at each call from the latest product count of the website.
Inputs: project_id (required). Returns: result
Create a matching job: find the same products on the reference website and the competitor websites. A paid job matches every pair of its websites, so the matches also connect the competitors with each other. All the websites must be scraped first. The job waits for approval with its price estimate; sample=True runs the free sample of the project instead. Approve it with scout_approve_job(job_kind="matching").
DeepScout matching does not need EAN codes. It finds the same product under different names, even in another language. Each uncertain pair gets a strict check with the product photos and data. A match must have the same brand, specifications, and pack size. It must also be the same variant, for example the same color, size, capacity, edition, or model number. Products with almost the same name often fail this check, so DeepScout matching is more accurate than homemade matching scripts.
The matches export has one row for each product and its match on another website. Join it with the product exports of the websites (website_id + product_id) to get:
assortment gaps: products that competitors sell and the reference website does not. These are competitor products with matches, but with no match on the reference.
price comparison of the market: the prices of one product on all the websites, also between two competitors.
assortment overlap: the products that each pair of websites shares. The completed job gives the counts in overlap.
The free sample matches a part of the reference website, once for each project. It does not match the competitors with each other. It shows the quality of the matches before the user pays. Run the sample before you create the paid job: a matching job that waits for approval blocks the sample.
Inputs: project_id (required), reference_website_id (required), competitor_website_ids (required), sample. Returns: result
Create a scraping job of a website: its full catalog, or only the given product URLs. The job waits for approval with its price estimate; nothing is charged yet. Approve it with scout_approve_job(job_kind="scraping"). Get the result of each URL of a job of given URLs with scout_get_url_results.
Inputs: project_id (required), website_id (required), urls. Returns: result
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
E-Commerce Intelligence MCP — 11 tools: price comparison, stock, reviews. 18 countries.
17 data tools in one MCP server, run on Apify: Google Shopping, Flights, Hotels, News, Images, Videos, Ads Transparency, Jobs and Trends; YouTube transcripts; AI visibility checks; Google Maps leads; App Store and Google Play reviews. Pay per result. Sign in with Apify (OAuth) or an API token.
Web scraping MCP server — scrape, extract structured data, screenshot any site with anti-bot bypass.
Clean, matched, AI-ready product data from any website
Add one secure layer between your agents and this server.
