The full upstream README, mirrored here for reference. Install config, tool schemas, adoption signals, and an original overview live on the Agentsync (remote) listing page.
Part of the xylem stack.
A Cloudflare Worker MCP server that makes claude.ai on your phone a peer in
the agentsync coordination mesh —
the same claims.json, the same agentsync branch, the same overlap and
compare-and-swap rules as local agentsync, with no git and no local clone.
Your laptop's Claude Code (local agentsync) and your phone's claude.ai
(agentsync-remote) claim work against the one shared claims.json. Two
transports, one mesh. A local peer and this remote peer are indistinguishable in
claims.json except by their agent id.
Sibling, not a fork. This is a second transport onto the same coordination file as local agentsync — not a variant of it. See
DESIGN.mdfor howgit push-as-CAS becomes GitHub-contents-API-as-CAS.
Seven tools over Streamable HTTP MCP:
| Tool | What it does |
|---|---|
survey | The whole board: every peer's claim, your conflicts, the mailbox. |
claim | Claim work; blocked if it overlaps an active peer (CAS-safe). |
check_conflicts | Re-check your claim against peers, optionally on one branch. |
update_status | Move your claim through planning / in-progress / done. |
release | Mark your claim done (done never blocks peers). |
history | Recent commits on the coordination branch (local + remote interleaved). |
mailbox | Human-in-the-loop notes: ask from the desktop, answer from the phone. |
Self-host this Worker in your own Cloudflare account. Every step is a click path — you never touch a terminal.
Click the button above. Cloudflare will:
agentsync-remote into your GitHub account,There are no databases or other resources to provision — this Worker keeps no state of its own (everything lives in your GitHub repo), so the deploy is just the Worker itself.
During the deploy dialog Cloudflare shows the Worker's variables. Set:
| Variable | Set it to |
|---|---|
REPO | Your coordination repo, as owner/name (the repo whose agentsync branch will hold claims.json). This is the one you must change. |
AGENT_ID | Leave as jonny-mobile, or pick an id for this peer. |
BRANCH | Leave as agentsync unless you want a different coordination branch. |
CLAIMS_PATH | Leave as claims.json. |
Finish the deploy. (If you skipped setting REPO here, you can set it later in
the dashboard — see below.)
The Worker reads two secrets. These are not part of the deploy dialog — you add them once in the dashboard after the first deploy.
First, make the GitHub token (this is the GH_PAT value):
GitHub → Settings → Developer settings → Fine-grained personal access tokens → Generate new token:
REPO).Then add both secrets to the Worker:
Cloudflare dashboard → Workers & Pages → your Worker → Settings → Variables and Secrets → Add → type Secret → add each, then Deploy:
| Secret | Value |
|---|---|
AUTH_TOKEN | A long random string you invent. It locks the endpoint — treat it like a password. |
GH_PAT | The fine-grained GitHub token you just made. |
While you're on this screen, confirm the REPO variable points at your
coordination repo (set it here if you skipped it in the deploy dialog).
Until
AUTH_TOKENis set the Worker answers every request with404(it fails closed). UntilGH_PATis set the tools return a clear error naming the missing secret. This is by design — an unconfigured endpoint looks like it doesn't exist.
Find your Worker's URL: Workers & Pages → your Worker shows it, in the form
https://<your-worker-name>.<your-subdomain>.workers.dev.
claude.ai (web) → Settings → Connectors → Add custom connector. Paste your
Worker URL with /mcp/ and your AUTH_TOKEN appended:
The token in the path is the auth — there is no separate login.
Ask Claude: "call survey". You should get the coordination board back (empty
claims on a fresh mesh — the Worker bootstraps the branch and file for you, so
a brand-new empty repo needs no manual setup).
The connector URL embeds your AUTH_TOKEN in the path
(…/mcp/<AUTH_TOKEN>). Anyone who has that URL can call your Worker and
read/write your coordination file. Treat the whole URL like a password:
AUTH_TOKEN in Settings → Variables and Secrets and redeploy; every old
…/mcp/<old-token> URL immediately returns 404. Update the connector in
claude.ai with the new URL.GH_PAT is likewise a credential — scope it to only your coordination repo
with only Contents: Read and write, so a leak can't reach anything else.
This is a deliberate design choice, not an oversight. claude.ai custom
connectors do not reliably send custom headers, so an Authorization: header —
the obvious alternative — cannot be depended on. Putting the credential in the
path is what makes the connector work at all.
Be clear about the price, because it is not the same as a header:
/mcp/***. The leak surface is everything around the Worker, which is
exactly what you cannot audit.Practical guidance:
AUTH_TOKEN,
update the connector.openssl rand -hex 32). The
comparison is constant-time in both content and length, so length is not
observable — but entropy is still your only defence against guessing.This is the whole point. The laptop and the phone coordinate through one file.
On the PC (local agentsync, Claude Code):
On the phone (claude.ai + agentsync-remote):
Back on the PC, the desktop agent (or you) reads the note via survey /
mailbox, answers it, and releases:
⚠️ Known gap: the mailbox is currently WRITE-ONLY across tiers
The "Back on the PC" step above does not work yet. This Worker's
mailbox()writes a top-levelnotes[]array intoclaims.json, and the local Python agentsync server has no concept ofnotes— it exposes nomailboxtool, and itssurvey()returns onlyme,branch,partnersandstale_claims. A note posted from the phone lands in the file correctly and is simply never surfaced to a desktop peer.That matters more than a missing feature normally would, because the repository-level agent instructions tell agents to raise judgment calls through this mailbox. An escape hatch that silently swallows the question is worse than no escape hatch: the agent believes it has asked and waits, or proceeds, on a question nobody will ever see.
Until the local side lands, read the mailbox from the remote peer (
mailboxwith nomessage, orsurvey), or readclaims.jsonon the coordination branch directly.The local-side change required (in the Python
agentsyncpackage — a separate repo, deliberately not modified here):
- Parse and preserve
notes. The claims-file reader must round-trip the top-levelnoteskey. Today an unrecognised key risks being dropped on the next local write, which would delete remote peers' notes. Preserving unknown top-level keys is the minimum safe change and should land first, independently.- Surface notes in
survey()— add anotesfield alongsidepartners/stale_claims, defaulting to[].- Add a local
mailbox(message=None, to=None)tool mirroring this one: append{from, to, message, at}under the same compare-and-swap the local claim writes already use, and cap retention at the 200 most recent notes (this Worker evicts oldest-first at that bound — the two sides must agree, or they will fight over the file).The on-disk shape is already compatible in both directions: this Worker only emits
noteswhen non-empty, so aclaims.jsonwith no mailbox activity stays byte-shape-identical to pure-local output.
On the phone again:
Every one of those steps was a compare-and-swap against the same
claims.json on the same branch. The laptop never saw a remote peer; the
phone never saw a git repo. Same mesh.
Non-secret config lives in wrangler.toml. Every var ships a
default so the Deploy button works with no edits; change REPO to your own repo.
| Var | Default | Meaning |
|---|---|---|
REPO | jarmstrong158/agentsync-remote | owner/name of the coordination repo. |
AGENT_ID | jonny-mobile | This peer's id in claims.json. |
BRANCH | agentsync | The coordination branch. |
CLAIMS_PATH | claims.json | The coordination file. |
Two secrets are set in the dashboard, never in the repo: AUTH_TOKEN (locks
the endpoint) and GH_PAT (GitHub Contents read/write on the coordination repo).
Point local agentsync and this Worker at the same REPO + BRANCH +
CLAIMS_PATH and they share one mesh.
Coordination correctness under contention is measured, not asserted — the full
write-up is docs/RELIABILITY.md, reproducible with the
commands below.
test/stress-cas.test.ts, seeded
0x5eed): 0 lost claims, 0 double-grants, and every race hit a real 409
retry — so the number isn't inflated by trivially-serialized runs.AUTH_TOKEN, and non-/mcp
paths all 404; a missing GH_PAT surfaces a named error, not a silent failure.initialize makes no network call, so it
answers instantly even on a cold isolate; it negotiates the client's requested
protocol version rather than rejecting a slightly-ahead client; and an
unexpected throw anywhere becomes a well-formed JSON-RPC error, never a bare
500 a reconnecting client would read as a hard failure.request / auth / handshake / tool_call / error, with the path
token always redacted), so a dropped handshake or a slow tool call is one
filter away in Cloudflare Workers Logs ([observability] is on in
wrangler.toml).Honest caveat: the thousand races are simulated in-process — GitHub's 409
compare-and-swap is reproduced by a fetch-mocked fake (test/helpers.ts),
not exercised over the live API. The results validate the Worker's coordination
logic — which is what runs in production — not GitHub's API, the network, or
real-world latency.
The canonical repo (jarmstrong158/agentsync-remote) deploys via GitHub Actions
→ Wrangler (.github/workflows/deploy.yml):
every push to main runs the suite and only deploys if it's green. That workflow
is guarded to the canonical repo, so a fork you created with the Deploy button
never tries to run it — your fork redeploys through Workers Builds instead.
Local checks (no Cloudflare credentials needed):
DESIGN.md — the CAS translation, overlap semantics, and why
the MCP handler is stateless and hand-rolled.docs/RELIABILITY.md — the measured reliability
report: 49/49 tests, five CAS race scenarios, and 1000 seeded contention races.