Rebuild public websites from evidence snapshots with website-rebuild-skill
Use website-rebuild-skill when an agent needs to classify a public site, capture a read-only evidence mirror, reconstruct behavior from shipped assets, and verify the rebuild with console, network, DOM, geometry, and pixel-diff gates.
npx skills add agentskillexchange/skills --skill rebuild-public-websites-from-evidence-snapshots-with-website-rebuild-skill
## What the agent/operator does
What this skill actually does
Use website-rebuild-skill to run a disciplined website reconstruction workflow. The agent takes a URL, grades feasibility, captures a read-only mirror with hashes and closure checks, traces minified source back into a human-readable project, and verifies the result against the source snapshot with automated acceptance gates.
The workflow is evidence-led: source snapshots stay immutable, every intentional difference is registered, and acceptance relies on console, network, DOM, geometry, and pixel comparison rather than a visual guess. The upstream repository documents support for Agent Skills-compatible runtimes and includes validated Claude Code and Codex runs.
## When to invoke
Invoke this skill when a user needs a private, auditable reconstruction of a public website for analysis, archival recovery, migration planning, or engineering study, and when the agent must prove what matches and what differs. It is not for quick redesign mockups or normal website building from a brief.
## Scope boundary
This is not a generic scraping library, site-cloning product, or deployment shortcut. The approved scope is classification, evidence capture, reconstruction, verification, and handoff. Keep legal and copyright decisions with the user; default outputs should remain private, noindex, and undeployed unless the user has the rights and explicitly chooses otherwise.
## Source
Upstream: https://github.com/boyang-hu/website-rebuild-skill
Documentation: https://github.com/boyang-hu/website-rebuild-skill