Logo
Tutorials

Extract and Check Proxies

Scrape proxy candidates from a source and send them directly into Proxy Checker.

This tutorial shows the first full workflow: collect candidates, normalize them, dedupe them, and check which ones are alive.

Demo Input

Use documentation-only examples when testing the flow:

https://example.invalid/proxies.txt
https://example.invalid/proxies.json
https://example.invalid/proxies.html

For real work, replace these with sources you are allowed to use.

1. Create the Extractor Task

  1. Open Core Tools -> Scrape Proxies.
  2. Choose Many as the source kind.
  3. Paste your source URLs.
  4. Set the task name to Docs extract demo.
  5. Set Parser profile to Auto.
  6. Set Dedupe mode to Host + port.
  7. Enable Prefer strongest if your sources mix schemes or credentials.
  8. Keep redirects enabled unless the source blocks them.
  9. Start the task.

2. Review Source Metrics

Check these metrics before trusting the output:

MetricGood signProblem sign
HTTP status200 or expected redirect403, 404, 5xx
Parser usedMatches the source typeAuto fell back unexpectedly
Candidate countHigher than final proxy countZero candidates
Duplicate countSome duplicates are normalVery high duplicates from noisy source
ErrorEmptyFetch, selector, or JSON path error

3. Send To Checker

  1. Filter out obviously bad source rows if needed.
  2. Use the handoff action to send extracted proxies to Check Proxies.
  3. Choose Visible rows if you filtered first.
  4. Confirm the new checker task opens.

4. Configure Checker

SettingValue
Protocol hintAuto
HTTPS checkEnabled
Expected status200
Resolve countryEnabled
Timeout10 to 20 seconds

5. Save Results

After the checker task finishes:

  1. Filter to alive rows.
  2. Prefer rows with HTTPS support.
  3. Review quality, anonymity, and latency.
  4. Save selected rows to a pool named Extracted checked pool.

You now have the first useful artifact: a saved pool of proxies that survived extraction and checking.

On this page