How ChatGPT handles this
You are in a ChatGPT conversation with the xSpeed app added, and you have a site where embedded videos slow every post down. You write what you see: the blog loads slowly when a post has a YouTube embed. ChatGPT calls list_sites, then list_modules to find the lazy loading module, then get_settings so it can tell you which parts of lazy loading are on today, such as images, embeds and click-to-play videos, before it suggests anything.
What comes back is an explanation in ordinary language and one concrete proposal: turn on Lazy-load embeds for blog, leave images as they are. The Hub tells the agent to briefly confirm intent before update_settings, so ChatGPT states the change and waits for your yes. After that it calls update_settings and reports the stored result. If any key or value breaks the module's rules, the whole call is refused and nothing is stored, a Pro-only module is refused on a site without a licence, and credential fields are refused over the Hub.
What differs in ChatGPT is how it decides to ask. It confirms write actions by default and treats a tool with no read-only annotation as a write action. The Hub's tools declare no such annotation, so expect a confirmation on list_modules and get_settings as well as on update_settings. You can expand a call to read the JSON it will send, and you can have ChatGPT remember an approval for a tool for the rest of the conversation, after which that tool runs without a prompt. The Hub runs a write as soon as the connection allows it. ChatGPT signs in with OAuth and gets the scopes it asks for, and the Hub's Read-only everywhere switch applies only to the connection token, which ChatGPT cannot send. For read-only access, sign in as a Viewer member of the workspace, or keep ChatGPT's own confirmations on.
Set up ChatGPT once
Already connected? Skip to the prompts. Alternatives and troubleshooting are on the ChatGPT 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.
2Turn on developer mode
In ChatGPT on the web, open Settings, then Security and login, and turn on Developer mode. On a Business workspace only admins and owners can use it. On Enterprise and Edu an admin grants access first, and you then turn it on under Settings, Apps, Advanced settings.
Settings > Security and login > Developer mode
3Create the app from the Hub URL
Open ChatGPT Plugins, select the plus button and create a developer-mode app. Name it xspeedhub, paste the MCP server URL and choose OAuth. Click Scan Tools, approve access on the xSpeed Hub page that opens, wait for the scan to list the tools, then click Create. Workspace admins can also start from Workspace settings, Apps, Create.
https://app.xspeedcache.com/xspeed/mcp
4Select the app in a chat
The app appears under Drafts in your app settings with a Dev label. In a new chat, open the plus menu in the composer, choose Developer mode and select xspeedhub. In a workspace, only an admin or owner can publish the app for other members.
+ > Developer mode > xspeedhub
Full ChatGPT 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. ChatGPT asks you to confirm write actions by default. You can expand a tool call to read the JSON it will send, and you can choose to remember an approve or deny choice for a tool for the rest of that conversation. ChatGPT respects a tool's read-only annotation and treats a tool without one as a write action; xSpeed Hub's tool definitions do not declare that annotation today, so expect confirmations on reads as well. OpenAI says confirmation also depends on app permissions and context, and some especially risky actions can be blocked. xSpeed Hub has no confirmation step of its own, so once you tell ChatGPT to remember Approve for a tool, that tool runs without a prompt for the rest of the conversation, as soon as your connection allows writes.
What do I ask?
Three prompts written for ChatGPT. More for this job are below.
Using xSpeed Hub, look at the lazy loading settings on blog and explain, in plain words, what each one does for a post with a YouTube embed.
Use xSpeed Hub to turn on lazy loading for embeds on blog only, and tell me the setting before and after.
Ask xSpeed Hub which modules on shop have settings I have never changed. Do not change anything yet.
What happens, step by step
Every xSpeed feature is a module with its own settings. Through xSpeed Hub an agent can list the modules a site has, read one module's settings, change them with a checked update, and turn page caching on or off. The site validates every write against the module's own schema, so a wrong key or value is refused with a reason instead of being quietly ignored.
01
List the modules
list_modules returns every Free and Pro module available on one site, with its settings schema and status. It tells the agent which module slugs exist, such as minify or gzip, and what each setting accepts.
02
Read the current settings
get_settings takes a module slug and returns its settings. Credential fields come back masked, so an agent cannot read a stored API token or password back in plain text.
03
Change them
update_settings takes the module and a values object of setting keys. The site checks the whole object first. If any key is unknown, any value is outside the schema, or a field is pinned by a constant in wp-config.php, it refuses the write, writes nothing and says which keys and why, sometimes with a "did you mean" hint.
04
Turn page caching on or off
Page caching has its own tool, toggle_cache, because it installs or removes the cache drop-in and the WP_CACHE constant. Asking update_settings to set cache_enabled does not work and is refused. If another caching plugin owns the drop-in, the site declines to enable xSpeed and returns blocked: true with the reason.
05
Check the effect
The result of a settings write is the stored settings. A module can add a warning when a setting is saved but has no effect on this server, for example Brotli on a server without the module for it, and the agent should pass that on. run_benchmark then times cached against uncached to see whether the change helped.
06
Apply to many sites only after reading one
update_settings and toggle_cache accept site: "all" or a list of handles. Each site validates against its own schema and answers in its own row, so a module missing on one site fails that row and the others still apply. Read one site first, then apply.
Reference
| Discover | list_modules (read): Free and Pro modules, settings schema, status, per site |
|---|---|
| Read settings | get_settings (read): argument module; credential fields masked |
| Change settings | update_settings (write): arguments module and values; validated by the site |
| Bad input | The whole write is refused and nothing is written; the error names the keys and reasons |
| Page cache on or off | toggle_cache (write): argument enabled; cache_enabled in update_settings is refused |
| Blocked by another plugin | toggle_cache returns blocked: true with blocked_reason when another plugin owns the cache drop-in |
| Every site at once | update_settings and toggle_cache accept site: "all" or a list of handles, one result per site |
| Pro modules | A Pro module on a site without an active licence cannot be read or written; the site refuses |
| Credential fields | Refused by default over MCP; set them in the xSpeed dashboard, not through the agent |
| Read-only connection | The connection token with Read-only everywhere on, or a Viewer member's sign-in: update_settings and toggle_cache are refused; list_modules and get_settings still work. An OAuth sign-in gets the scopes the client asks for, read and write by default, clamped to the member's role |
| Confirmation | None on the Hub; the Hub tells the agent to briefly confirm intent for these two, and your client's prompt is the gate |
Rules worth keeping
- Read before you write. Ask for list_modules and get_settings first, so the agent changes real keys with valid values instead of guessing.
- update_settings and toggle_cache change how a live site behaves. They run as soon as your connection allows writes. The Hub tells the agent to briefly confirm intent first, which is guidance to the agent, 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. The switch does not limit an OAuth sign-in.
- For site: "all", the Hub tells the agent to confirm once with "this will touch all N sites", then make one call. The result is per site, and some sites failing is a partial result to report, not a blanket success or failure.
- Do not ask an agent to set credentials such as a Cloudflare token or Redis password. The site refuses those writes by default, and they belong in the xSpeed dashboard.
- If a settings result says a value is saved but inactive on this server, relay that. Do not tell the user the feature is enabled.
Good to know with ChatGPT
ChatGPT needs developer mode on the web to add the Hub as a custom MCP app, and the mobile app cannot use it. OpenAI's pages differ on plans: the developer-mode guide lists Pro, Plus, Business, Enterprise and Edu, while the help center calls full MCP with write actions a beta for Business, Enterprise and Edu and says Pro connects with read and fetch permissions. If update_settings and toggle_cache are missing, your plan is the likely reason. Ask ChatGPT for the exact change and apply it in the xSpeed dashboard.
More prompts for this job
They work in any client connected to xSpeed Hub.
Use xSpeed Hub to list the modules on shop and tell me which ones are off.
Ask xSpeed Hub for the current minify settings on blog.
With xSpeed Hub, turn on HTML minification on blog, then benchmark the cache and tell me if it helped.
Use xSpeed Hub to turn page caching off on staging while I debug, and tell me when it is off.
Using xSpeed Hub, read the gzip settings on shop first, then enable the same options on every site.
Use xSpeed Hub to set the page cache lifespan on docs to 30 days, and if the site refuses it, tell me why.
Frequently asked questions
Keep going
Tune cache settings with other agents
More with ChatGPT
Documentation
- How to enable Page Cache
- How to minify CSS and JavaScript
- How to add cache rules and bypasses
- Site MCP tool reference
- How to write prompts for xSpeed Hub
- ChatGPT + xSpeed
- Every AI agent that works with xSpeed
From the blog