Fetching from the wire…
Public story · 2026-08-17 · high
The project has 21 Show HN points and publishes no reliability numbers, making the pattern worth copying even without adopting the library.
Why now: PyScrappy surfaced in the 2026-08-17 briefing's Show HN sweep, still small and unproven beyond that post.
PyScrappy combines self-healing scraper selectors with an MCP server that lets an agent drive the scraper directly, per its GitHub listing. That saves teams running agentic scrapers from hand-editing XPaths every time a target site redesigns its DOM.
A selector normally breaks the moment a page's DOM shifts, and someone has to open the code and rewrite the XPath by hand. PyScrappy wires the fix into the selector itself, so the model recovers without a person touching the code.
The project surfaced as a Show HN post with 21 points, and PyScrappy doesn't publish reliability numbers for how often the self-healing actually works. That's thin evidence for a production bet.
The library matters less than the shape: self-healing selectors paired with an MCP server, so an agent controls scraping instead of generating throwaway scripts. Teams that skip this pattern will keep burning hours rewriting selectors by hand. That holds whether or not PyScrappy itself survives as a project.
Each link below shares sources, entities, or timing with this story.
Cursor uses MCP / Shared entities / Same source domain / Shared topic / Earlier coverage
Linked by a graph relationship (Cursor uses MCP); both cover DOM, GitHub, MCP; reported by the same outlet (github.com).
Claude Code uses MCP / Shared entities / Same source domain / Shared topic / Earlier coverage
Linked by a graph relationship (Claude Code uses MCP); both cover GitHub, MCP, Self; reported by the same outlet (github.com).
Linked by a graph relationship (Claude Code uses MCP); both cover GitHub, MCP; reported by the same outlet (github.com).
Claude Code uses MCP / Shared entities / Same source domain / Shared topic / Earlier coverage / Tension
Linked by a graph relationship (Claude Code uses MCP); both cover GitHub, MCP; reported by the same outlet (github.com).
Codex CLI uses MCP / Shared entities / Same source domain / Shared topic / Earlier coverage / Tension
Linked by a graph relationship (Codex CLI uses MCP); both cover GitHub, MCP; reported by the same outlet (github.com).
Claude uses MCP / Shared entities / Same source domain / Shared topic / Earlier coverage / Tension
Linked by a graph relationship (Claude uses MCP); both cover GitHub, MCP; reported by the same outlet (github.com).
Cursor uses MCP / Shared entities / Same source domain / Shared topic / Earlier coverage
Linked by a graph relationship (Cursor uses MCP); both cover GitHub, MCP; reported by the same outlet (github.com).
Claude Code uses MCP / Shared entities / Same source domain / Shared topic / Earlier coverage
Linked by a graph relationship (Claude Code uses MCP); both cover GitHub, MCP; reported by the same outlet (github.com).