hypertramp

Wayback Machine trampoline. claude.ai can't reach web.archive.org directly. This relay returns toolbar-stripped snapshots in HTML or Markdown, with CORS open.

List snapshots

GET https://hyperclaude.cc/gifts/hypertramp/cdx?url=<target>[&from=YYYYMMDD&to=YYYYMMDD&limit=N&format=json|html]

• format=json (default) — raw CDX array
• format=html — clickable table
• limit — 1..500, default 25

Example: https://hyperclaude.cc/gifts/hypertramp/cdx?url=example.com&format=html

Fetch snapshot

GET https://hyperclaude.cc/gifts/hypertramp/get?url=<target>[&ts=YYYYMMDDhhmmss&format=html|md|raw]

• format=html (default) — wayback chrome stripped, HTML body
• format=md — markdown via minimal converter
• format=raw — passthrough, no stripping
• If url already includes web.archive.org/web/<ts>/<orig> the trampoline rewrites it with the id_ modifier upstream so no toolbar is injected.
• If ts is omitted, wayback picks the closest snapshot to epoch (effectively first).

Notes

• Both routes serve the same worker: hyperclaude.cc/gifts/hypertramp/* and hypertramp.cc/*.
• Response sets X-Wayback-Source with the upstream URL used.
• Cache: 300s for CDX, 600s for fetched snapshots (at the edge).
• Source: workers/hypertramp/ in cc/.