Skip to main content

On This Page

Automated MTProto Proxy Scraper: Fresh Telegram Proxies via CI/CD

2 min read
Share

These articles are AI-generated summaries. Please check the original sources for full details.

Developer Yagami200 created a Python script to automate the discovery of working MTProto proxies for Telegram. The first run surfaced around 30 live proxies, all supporting fake TLS, without any manual effort.

Why This Matters

Telegram proxy lists often contain dead links, requiring constant manual testing. This automated approach replaces painstaking human verification with scheduled CI/CD, ensuring a constantly fresh set of working proxies for bypassing restrictions or adding redundancy.

Key Insights

  • The script scrapes multiple public proxy channels and tests each proxy for liveness: a process that used to be done manually.
  • It outputs clean JSON and a live web page, making the filtered proxy list immediately usable without additional parsing.
  • The scraper runs on a schedule via GitHub Actions, eliminating the need for a dedicated server or manual scheduling.
  • Around 30 working proxies were identified on the first pass: demonstrating the viability of automated proxy verification.
  • The repository is public and self-hostable: allowing others to run their own proxy verification or simply use the published list.

Practical Applications

  • Bypassing regional restrictions: Use the generated proxy list in Telegram clients to access blocked services without manual proxy hunting.
  • Adding connection redundancy: Automatically rotate proxies from the JSON output to maintain stable access when individual proxies fail.
  • Self-hosting the scanner: Fork the repo to add custom proxy sources or modify testing parameters for specialized use cases.
  • Pitfall: Relying solely on public proxy lists without validation leads to latency spikes and connection drops; this script mitigates that by pre-testing each proxy.

References:

Continue reading

Next article

LangChain Agent Silently Failed for 2 Weeks, Costing $2,400: Why Trace Observability Misses Semantic Errors

Related Content