Manage every WordPress site at once with OpenClaw

Managing every WordPress site at once with OpenClaw means having an always-on agent call xSpeed Hub across your whole workspace, on request or on a schedule, and report what each site did.

list_siteslist_workspacespurge_cachetoggle_cacheupdate_settingsstart_preloaderstop_preloaderoptimize_siterun_benchmark

Last updated: October 2026

Example session · OpenClaw

You: Use xSpeed Hub to benchmark every site now, and tell me about any that failed.

list_workspaces read-only
→ 2 workspaces; default has 13 sites

run_benchmark(site: all) read-only
→ 12 benchmarked; client-m unreachable

OpenClaw: Twelve of thirteen sites measured. client-m did not respond. This was a read-only run, so nothing on any site was changed.

How OpenClaw handles this

OpenClaw differs from the editor and terminal agents because it keeps running. That suits an agency owner who wants a standing check on the whole portfolio, not only a command after a deploy. You can ask it right now to benchmark all sites, or set an automation to do that every morning. It calls run_benchmark with site set to "all", the Hub runs the sites one at a time with a pause of one second, and you get a result for each site.

The Hub has no scheduler for agent prompts. The timer lives in OpenClaw, and the Hub only answers the calls. Of the seven fan-out tools, run_benchmark is a read tool, so a scheduled benchmark across every site changes nothing. A scheduled purge, toggle or settings change is another matter: purge_cache, toggle_cache, update_settings, start_preloader, stop_preloader and optimize_site all write, and nobody is there to answer a prompt. The Hub's confirm-once instruction names purge, toggle, settings and preloader calls but not optimize_site, and it needs a person in the conversation either way.

So decide what each automation is allowed to do before you schedule it. A report that reads, compares and flags a drop is safe to leave alone. A job that clears caches or flips settings should wait for you, or run on a connection you have deliberately given write scope. Per-site results still arrive for every run, so a partial failure is visible in the next report instead of buried.

Set up OpenClaw once

Already connected? Skip to the prompts. Alternatives and troubleshooting are on the OpenClaw guide.

1Put your sites in xSpeed Hub

Sign in at app.xspeedcache.com with Google or email; the Hub is free and has no site cap. Then connect each WordPress site from its own dashboard: click Connect Hub in the xSpeed Cache top bar, then Connect via xSpeed Hub. Each site needs the free xSpeed Cache plugin.

2Save xSpeed Hub as an MCP server

Run this on the machine that hosts your OpenClaw Gateway. Set the transport to streamable-http: OpenClaw falls back to SSE when you leave it out. With --auth oauth the command saves the server without a connection test, because sign-in has to happen first.

Terminal

openclaw mcp add xspeedhub \
  --url https://app.xspeedcache.com/xspeed/mcp \
  --transport streamable-http \
  --auth oauth

3Sign in with openclaw mcp login

OpenClaw prints an authorization URL and listens on a loopback address for the callback. Open the URL, sign in to xSpeed Hub and approve; the token exchange finishes by itself. There is no token to paste. If your browser is on another machine, copy the code from the redirect and pass it back with openclaw mcp login xspeedhub --code. Then check the connection.

Terminal

openclaw mcp login xspeedhub
openclaw mcp doctor xspeedhub --probe

1 / 3

Full OpenClaw setup, sign-in options and FAQ

Before you send a prompt that changes something

Reads change nothing on your sites, though contact_support emails xSpeed support. Writes run once your connection allows them, so your client's approval prompt and a read-only connection (the token's Read-only everywhere switch, or a Viewer sign-in) are the gates that matter. OpenClaw gives you two client-side gates, and the Hub adds none. First, openclaw mcp tools xspeedhub --include limits which tools exist for the agent, so a list of read tools keeps every write out of reach. Second, when the Codex runtime runs your agent turns, MCP tool approvals follow the session's permission posture, and the default full-permission posture does not prompt. openclaw mcp configure xspeedhub --approval prompt asks on every call, and Allow Always saves a grant for that one tool with any arguments. OpenClaw's docs describe that per-call prompt for the Codex runtime; with its own embedded runtime, use the tool filter to decide what the agent can call. The Hub has no confirmation step of its own.

What do I ask?

Three prompts written for OpenClaw. More for this job are below.

Prompt for OpenClaw

Every morning, use xSpeed Hub to benchmark all my sites and tell me which ones got slower than yesterday.

Prompt for OpenClaw

Use xSpeed Hub to run the benchmark on every site now and post one line per site.

Prompt for OpenClaw

Ask xSpeed Hub to list my workspaces and the site count in each, so I can choose which one to schedule against.

What happens, step by step

If you run many WordPress sites, repeating one action on each is the slow part. Through xSpeed Hub, seven tools accept every site in a workspace, or a list of handles, in a single call. The Hub runs the sites one after another and returns a result for each, so a site that fails is named instead of hidden. Everything else in the Hub works on one site per call.

See what you can reach

list_sites returns each site's handle, URL, status and scopes. When you have more than one workspace, list_workspaces returns each one's handle, id, your role, its site count, which one is the default, and whether this connection can reach it.

Choose the scope of the call

The agent passes site: "all" for every site in one workspace, or an array of handles for exactly those sites. For a workspace other than the default it also passes the workspace argument. Omit workspace and the call acts on the default workspace.

Confirm once

Before a purge, toggle, settings change or preloader call that touches every site, the Hub tells the agent to say "this will touch all N sites" and wait for you, then make one call. That instruction does not name optimize_site, so for it your client's prompt is the gate. It is guidance to the agent, not a lock. It is one tool call, so one approval in your client covers every site in it.

Run it as one call

The Hub runs the sites one after another with a one-second pause between live calls, so a fleet purge is not a burst of traffic at your servers. Revoked sites are skipped from "all". Each site's action lands in the activity log as its own row.

Read the result per site

The reply is a summary of targets, succeeded and failed, plus one row per site with ok and either its data or its error. Some sites failing is a partial success. Only a run where every site failed comes back as an error.

Follow up on the failures

The agent reports which sites failed and why, then retries just those handles with an array. Tools that read one site, such as get_cache_status and get_health, do not fan out, so a fleet-wide check means one call per site.

Reference

Fan-out tools	purge_cache, toggle_cache, update_settings, start_preloader, stop_preloader, optimize_site, run_benchmark
Selector	site: "all", or an array of handles
What all covers	Every site this connection can see in one workspace, minus revoked ones
Pacing	One after another, with a 1 second pause between live calls
Result	summary (targets, succeeded, failed) and a row per site: ok, then data or error
Partial failure	Reported per site; the call counts as an error only when every site failed
Workspace argument	Optional. Omit it for the default workspace; name one by its handle or id from list_workspaces
Reach	Fixed when you approve the connection; list_workspaces shows reachable: false for the rest
Not fan-out	get_health, get_cache_status and every other tool take one site per call
Slow tools	optimize_site takes up to 2 minutes per site; a target_score run is capped at 12 rounds per call
Read-only	The connection token with Read-only everywhere on, or a Viewer sign-in, refuses every fan-out write; run_benchmark works. An OAuth sign-in gets the scopes the client asks for
Hub daily pass	Re-verifies sites, snapshots cache status and hit ratio, reads stored reports, alerts by Slack webhook or email

Rules worth keeping

Confirm once, then call once. The Hub has no confirmation step of its own, so your client's approval prompt and a read-only connection (the connection token with Read-only everywhere on, or a Viewer sign-in) are the gates that matter. The Hub's "this will touch all N sites" is an instruction to the agent for purge, toggle, settings and preloader calls, and it does not name optimize_site.
"all" is one workspace, never every workspace. To act in another workspace, name it. A write to a non-default workspace must name it, and an array of handles is looked up only in the workspace the call names.
Read every result per site. Treat some sites failing as a partial result, and never accept a blanket "done" from the agent.
Try a behaviour change on one site before you send it to the fleet. toggle_cache and update_settings on every site are one call, and each site checks values against its own settings, so a Pro setting can fail on a Free site.
optimize_site on many sites is slow, because each site can take up to two minutes inside one call. Use dry_run first, or name a short list of handles.
The Hub has no scheduler for agent prompts. Its own daily pass and alerts run without an agent, and an agent can only act when you send a prompt or your AI client runs one on a timer.

Good to know with OpenClaw

An automation that writes has no one to ask. A scheduled run is unattended, and the Hub's instruction to confirm before a fleet-wide purge, toggle, settings change or preloader call only works when a person is in the conversation. OpenClaw gives you two gates. A tool filter on the server decides which Hub tools exist for the agent, so a list of read tools leaves nothing to purge with. A per-call prompt exists when the Codex runtime runs your turns, but OpenClaw's default full-permission posture does not prompt, and Allow Always saves a grant for one tool with any arguments, site: "all" included. Keep scheduled fleet jobs to run_benchmark and other reads, filter the write tools out, and give any automation that writes its own connection with a scope you chose.

More prompts for this job

They work in any client connected to xSpeed Hub.

Prompt

Use xSpeed Hub to list every site I can reach, and every workspace, and say which workspace you act in by default.

Prompt

With xSpeed Hub, purge the cache on every site in my workspace and tell me which ones failed.

Prompt

Use xSpeed Hub to run a benchmark on all my sites and rank them by cached response time.

Prompt

Use xSpeed Hub to start the preloader on blog, shop and docs only.

Prompt

With xSpeed Hub, turn page caching on for every site where it is off. Check each site first.

Prompt

Use xSpeed Hub to stop the preloader everywhere. The servers are busy.

Prompt

Ask xSpeed Hub to preview a safe optimize_site pass on every site with dry_run and summarize what it would change.

Prompt

Use xSpeed Hub to purge every site in the client-acme workspace, not my default one.

Frequently asked questions

Yes. The Hub has no scheduler of its own, but OpenClaw can call it on a timer. A benchmark across all sites is a read and changes nothing, which makes it a good scheduled job.

Only if you accept that nobody is there to confirm it. The Hub runs a write as soon as the connection allows writes, and a scheduled run has no person to answer. Filter the write tools out of the server, use a connection scope you chose on purpose, or leave writes to manual runs.

Every site the connection can see in one workspace. It never crosses workspaces, and revoked sites are skipped. If your connection is limited to some sites, "all" quietly means the sites you are allowed, with no error naming the rest. To act in another workspace, the agent names it with the workspace argument.

The other sites still run. The reply has a summary and a row for each site with ok and its data or its error, so a site that was unreachable is named. A run counts as an error only when every site failed. Ask the agent to retry just the failed handles.

No. get_health and get_cache_status take one site per call, so the agent loops. The tools that fan out are purge_cache, toggle_cache, update_settings, start_preloader, stop_preloader, optimize_site and run_benchmark. The Hub dashboard has a fleet overview for health.

Yes, on its own. A daily pass re-verifies each site, retrying a failed probe once, snapshots cache status and hit ratio, and reads the PageSpeed and GTmetrix reports your sites already stored. It can alert you by Slack webhook or email when a site becomes unreachable, caching is switched off, the hit ratio falls below a floor you set, or a score regresses. Each alert fires once per incident, not every day.

The Hub has no confirmation step. Its instructions tell the agent to confirm once with you, saying how many sites a call will touch, but that is guidance and not a lock. Your AI client's approval prompt and a read-only connection (the connection token with Read-only everywhere on, or a Viewer sign-in) are what stop a write. The instruction covers purge, toggle, settings and preloader calls, not optimize_site. The dashboard's own Purge all button does confirm before it runs.

No. It cannot add, remove or rename sites, change members, roles or billing, mint or rotate tokens, or widen the workspaces this connection reaches. Those stay in the Hub dashboard.

Manage every site at once, from OpenClaw.

xSpeed Hub is free and has no site cap. Connect your sites once and ask from the agent you already use.

Open xSpeed Hub OpenClaw setup

Manage every WordPress site at once with OpenClaw

How OpenClaw handles this

Set up OpenClaw once

What do I ask?

What happens, step by step

Reference

Rules worth keeping

More prompts for this job

Frequently asked questions

Keep going

Manage every site at once, from OpenClaw.