# Zyte demos self-healing Scrapy spiders that repair crawls after site redesigns

> Zyte published a blog post detailing two workshop demos from Extract Summit Austin where an agent fixes a Scrapy spider after a site redesign and heals a live crawl mid-run using Spidermon and the Scrapy MCP, including cost measurements and design mistakes.

- Canonical: https://extractfeed.io/story/zyte-demos-self-healing-scrapy-spiders-that-repair-crawls-af-9663d14/
- JSON: https://extractfeed.io/api/v1/stories/zyte-demos-self-healing-scrapy-spiders-that-repair-crawls-af-9663d14.json
- Beat: Agents & MCP · Evidence: Primary source · Type/significance: analysis/2 · First seen: 2026-10-23T00:05:46.691510+00:00 · Updated: 2026-10-23T00:05:46.691510+00:00 · Edition: 2026-11-05
- Framing: model-written (headline, standfirst, why it matters, tags); source facts deterministic

## Briefing
- Zyte: Zyte published a blog post detailing two workshop demos from Extract Summit Austin where an agent fixes a Scrapy spider after a site redesign and heals a live crawl mid-run using Spidermon and the Scrapy MCP, including cost measurements and design mistakes.

## Why it matters
This is a practical demonstration of agent-driven resilience in web scraping, showing how AI can reduce the manual maintenance burden of production crawlers. By quantifying costs and sharing design failures, Zyte provides a realistic benchmark for teams considering similar self-healing approaches, though the post is a tutorial writeup, not a product launch.

## Sources
- Primary source · Zyte · 2026-10-22 — [Self-healing spiders: letting an agent fix your Scrapy crawl, with a lid on it](https://www.zyte.com/blog/self-healing-spiders-agent-fix-scrapy-crawl/)

## Watch next
Will Zyte productize this self-healing agent pattern into a managed service or keep it as a reference implementation?

Topics: Zyte, Scrapy, Spidermon, Extract Summit Austin, self-healing, scrapy, spidermon, mcp, web-scraping, agent, crawl-maintenance

---
extractfeed is an agent-readable changefeed for web scraping and data extraction. Index: https://extractfeed.io/agents.md