curl -s https://www.pathwren.workers.dev/git/index.json # this page, as JSON
No key, no account, no handshake — every page here has a JSON twin one hop away. Machine doors: 6 keyless GET tools · documents.json · changes · llms.txt · openapi.json · agent card · mcp · a2a
The git repositories this host serves, how to clone them, and what a Go module proxy resolves here. Dumb HTTP, no forge, no account, read-only.
git clone https://www.pathwren.workers.dev/git/ai-crawler-index.git
| Repository | Branch | HEAD | Loose objects | Served |
|---|---|---|---|---|
| ai-crawler-index.git | main | 03256e336176 | 173 | 1,693 KB |
A whole clone is re-walked by playbooks/git_dumb_http_watch.py — last run 2026-09-05T21:10:09Z: 19 request(s), 0 failure(s)
Each repository is a bare one, written by git update-server-info
and served as ordinary static files over Git's dumb HTTP protocol. There is no
git daemon, no smart transport and no forge behind it: git clone,
git fetch and git ls-remote work, and nothing else does.
It is read-only for everybody — there is no push endpoint to refuse you.
Patches are welcome by mail as git format-patch output; the address is on
/contact.
cmd/go and proxy.golang.orgThis host publishes no Go module, and no go-import meta tag at
any address. A module path whose repository root ends in .git is
resolved by the go command's generic VCS rule with no meta tag at all, so
www.pathwren.workers.dev/c/go-module-proxy/git/ai-crawler-index.git
does name this repository to a module proxy — and proxy.golang.org has fetched it on
that path. What it will not do is build: the repository contains data, Python,
JavaScript and Ruby, and not one .go file, so go get ends at
“no Go files” and pkg.go.dev returns 404 for the same reason. It is an
archive in the module proxy, not a library.
Which is why nothing here serves a go-import tag: the tag would
claim a module root that does not exist, and send you to fetch a repository that
cannot compile. /c/go-module-proxy/git and /c — the two
addresses GoCommand/1 took a 404 on at 2026-09-05T23:47:40–41Z while
walking this path's prefixes — now answer with this page and with
/c, which is the honest “not a module root, and here is what
it is instead”.
AI Crawler Index — the source behind https://www.pathwren.workers.dev (independent, automated, CC0 data / MIT code) Code MIT, data CC0-1.0. The same data is published as /data/agents.json, as packages on PyPI, npm and RubyGems (/packages.html), and as this whole site.
Every path here is also served under /c/<channel>/, which is the same
bytes with a note about which directory sent you — see /c. Clone either
spelling; they are one repository.