# Billing Source: https://docs.wpos.ai/agencies/billing Plans, credits, and site licenses, and how they work together in WPOS. Billing in WPOS is built around three things: a **plan**, a pool of **credits**, and a set of **site licenses**. This page explains how they fit together. For current plan names and prices, the source of truth is the pricing page on [wpos.ai](https://wpos.ai) and your account at [app.wpos.ai](https://app.wpos.ai). This page describes how billing works in general. Exact names, limits, and prices live on the pricing page and in your account, which always win over anything written here. ## The three pieces Your subscription. Sets your monthly credits and how many sites you can operate before adding more. The unit of work. Running the agent spends credits. Your plan refills them each month. Each connected site uses one license. Plans include a number of them, and you can add more as your fleet grows. ## How credits work Credits are the unit WPOS bills work in. When the agent runs work on a site, it spends credits from your account balance. * Your plan tops your balance up to its monthly allotment on each billing cycle. * Usage is tracked per account, so you can see where credits go across every site you operate. * Heavier work spends more than light work. Building memory once with a [Playbook](/memory/playbook) and [Skills](/memory/skills) makes later work cheaper, because the agent repeats instead of re-deriving. Need more than your monthly allotment in a busy month? Credit packs are one-time top-ups that do not expire, so a heavy month does not force a permanent plan change. See the pricing page for pack sizes. ## Site licenses Every site you connect consumes one site license. Your plan includes a set number, and you add more when you connect more sites. Members on a site are not billed per seat, so bringing your team or a [Guest agent](/operate/guest-agent) onto a site does not change the cost. ## Managing billing Go to [app.wpos.ai](https://app.wpos.ai) and open billing. See your current credit balance and where it is going across connected sites. Change plan, add site licenses, or buy a credit pack as your fleet and workload grow. Running many sites and want a plan built for that? See agency pricing on [wpos.ai](https://wpos.ai), or contact [support@wpos.ai](mailto:support@wpos.ai). # Whitelabel Source: https://docs.wpos.ai/agencies/whitelabel Present WPOS under your own brand to clients. Whitelabel lets you put WPOS to work inside your agency offering while keeping the brand relationship yours. Your clients experience the result, smooth delivery and ongoing care, while the operating layer does the heavy lifting in the background. Whitelabel options depend on your plan. See [Billing](/agencies/billing) for what is included, or contact [support@wpos.ai](mailto:support@wpos.ai) to talk through agency setup. ## Why agencies whitelabel Agencies sell delivery and the confidence that a site is being looked after, not a tool subscription. Whitelabel keeps that promise intact. The client sees your service running reliably; they do not need to know which engine is underneath, and the relationship, the trust, and the billing stay with you. ## What you control The client deals with you. You own the account, the plan, and the billing, and you decide what the client sees and touches. Each client's [Playbook](/memory/playbook) and Branding kit keep their site on-brand automatically, so whitelabelled work still reflects that client. Use the [Guest agent](/operate/guest-agent) to give a client scoped access to their own site, without exposing your [Workspace](/agencies/workspace) or the rest of your fleet. Operate every client site from one [Workspace](/agencies/workspace), each with its own memory, while presenting each one individually to its owner. ## How it fits together Whitelabel is not a single switch, it is how the other surfaces combine when you operate on a client's behalf. Build a [Playbook](/memory/playbook) per client so every change matches their brand and rules without you restating them. Keep full operation to your team, or hand the client a scoped [Guest agent](/operate/guest-agent) for self-serve requests. Run all of it from the [Workspace](/agencies/workspace), with usage and billing consolidated under your account. The stronger each client's [Playbook](/memory/playbook), the more invisible the operating layer becomes. The site already knows the brand, so the output looks like your agency's work, not a generic tool's. Setting up whitelabel for an agency? Contact [support@wpos.ai](mailto:support@wpos.ai) and we will walk through the options on your plan. # Workspace Source: https://docs.wpos.ai/agencies/workspace Manage every site you operate from one cross-site view. The Workspace is the cross-site view at [app.wpos.ai](https://app.wpos.ai). Where the [Command Center](/operate/command-center) operates one site from inside wp-admin, the Workspace is where you see and manage your whole fleet. Keep the two straight. The Workspace is the cross-site dashboard at app.wpos.ai. The Command Center is the per-site surface inside each site's wp-admin. ## What you do in the Workspace * See every connected site in one place. * Move between sites and open the Command Center for any of them. * Manage connections, access, and the people on your team. * Track work across sites. ## Built for fleets Agencies do not run one site, they run many. The Workspace is built for that: connect each client site once, and operate them all from a single account, with shared [memory](/memory/overview) where it makes sense and per-site rules where it does not. Present WPOS under your own brand to clients. Plans, usage, and how billing works. # API reference Source: https://docs.wpos.ai/api-reference/introduction The WPOS API and how to generate an interactive reference from it. This section will hold the WPOS API reference. Once an OpenAPI specification is added to the repo, Mintlify generates an interactive playground for every endpoint automatically, so this section stays in sync with the API by design. This is a placeholder. To turn it into a full, interactive reference, add an OpenAPI spec and point `docs.json` at it. See the setup notes below. ## How to populate this section Drop an `openapi.json` or `openapi.yaml` file into the repo, for example at `api-reference/openapi.json`. Add the spec to the API reference tab in `docs.json`. Mintlify generates one page per endpoint, each with an interactive playground. Regenerate or update the spec whenever the API changes, and the reference updates with it. ## What developers get * An interactive playground for every endpoint. * Request and response examples generated from the spec. * A reference that an AI agent can read over [MCP](/connect/mcp) alongside the rest of these docs. # Connectors Source: https://docs.wpos.ai/connect/connectors Bring the tools your agency already uses into WPOS. Connectors let WPOS reach the tools your agency already runs on. Connect an account once, and WPOS can pull from and push to that tool as part of the work it does on your sites. ## What you can connect WPOS connects to over a thousand external tools, including Gmail, Slack, GitHub, Google Drive, Notion, and the rest of the stack a modern agency uses. The catalog is large and growing. ## What this unlocks * Pull content or assets from a connected tool straight into a build. * Push updates out, for example notifying a Slack channel when a Task finishes. * Let a [Skill](/memory/skills) reach across tools as part of a procedure. ## How it fits the flywheel Connectors are the outer ring of the [memory](/memory/overview) flywheel. They extend how far the work can reach, while Playbook, Skills, and Patterns govern how the work is done. Together they let WPOS operate WordPress in the context of your whole toolchain, not in isolation. Authorize the tool from your WPOS account. The connection is stored securely. Reference the connected tool in the [Command Center](/operate/command-center) or inside a [Skill](/memory/skills). Actions that send data out still pass through [Plan mode](/operate/plan-mode). Want your own agent to drive WPOS instead? See [MCP](/connect/mcp). # MCP Source: https://docs.wpos.ai/connect/mcp Operate WPOS from your own agent over the Model Context Protocol. WPOS is built on the Model Context Protocol (MCP). That means it is not just a product you use through a UI. It is also a set of tools your own agent can drive. ## Two ways MCP shows up Internally, WPOS exposes WordPress operations as MCP tools, which is how it builds widgets, edits content, and manages sites in a structured, reviewable way. These docs are also served as an MCP server, so your own Claude or Cursor can load them and operate WPOS with full knowledge of how it works. ## Using the docs MCP server This documentation is available to AI agents as structured Markdown and as an MCP server. Point your agent at it and it can answer questions about WPOS and help you operate it, citing these pages directly. Every page in these docs is also available as plain Markdown by adding `.md` to the URL, which keeps token usage low for any agent reading them. ## Why this matters for agencies If your team already works in Claude or Cursor, MCP means WPOS fits into that workflow rather than sitting beside it. The same memory that powers the [Command Center](/operate/command-center), your [Playbook](/memory/playbook), [Skills](/memory/skills), and [Patterns](/memory/patterns), is what your agent operates against. # Install the plugin Source: https://docs.wpos.ai/get-started/install Install WPOS on a WordPress site and connect it to your account. WPOS runs as a WordPress plugin paired with the WPOS backend. The plugin gives you the Command Center inside wp-admin and keeps the site in sync with your account. ## Requirements WordPress 6.0 or newer (WordPress 7.0 fully supported) PHP 7.4 or newer Ability to install plugins (administrator access) A WPOS account at [app.wpos.ai](https://app.wpos.ai) ## Install from the dashboard Add the WPOS plugin to your site, either from the WordPress plugin directory or by uploading the ZIP from your WPOS account. Activate the plugin. A WPOS menu item appears in the wp-admin sidebar. Open the WPOS panel and connect the site to your account. The site is now linked to your [Workspace](/agencies/workspace). ## Manual upload Download the latest plugin ZIP from your WPOS account. In wp-admin, go to Plugins, then Add New, then Upload Plugin, and select the ZIP. Activate, then open the WPOS panel and connect to your account. ## Updates WPOS ships updates through the normal WordPress update flow. When a new version is available, you will see it on the Plugins screen like any other plugin. Running many client sites? See [Workspace](/agencies/workspace) for managing updates and connections across your whole fleet from one place. ## Troubleshooting Confirm the plugin is active under Plugins, and that your user has administrator capabilities. Clear any full page cache and reload wp-admin. Check that the site can reach the WPOS backend over HTTPS and that no security plugin is blocking outbound requests. Re-open the WPOS panel and retry the connection. # What is WPOS Source: https://docs.wpos.ai/get-started/introduction WPOS is an operating layer for WordPress, not another AI site builder. ## The short version WPOS connects to a WordPress site, builds a memory of how that site is put together, and then runs the operational work on it: building widgets and pages, editing content, managing plugins and settings, and carrying out multi step tasks. It works inside wp-admin through the [Command Center](/operate/command-center) and across many sites through the [Workspace](/agencies/workspace). ## What WPOS is not WPOS is not an AI website builder that spits out a generic site and leaves you stuck inside a closed platform. The sites stay native WordPress. The point is not to generate a site once. The point is to operate sites continuously, with context that survives between sessions. The enemy is forgetting. Most AI tools start from zero every conversation. WPOS remembers your site, your client's brand rules, and the way your team does the work, so the second task is faster than the first and the hundredth is faster than the second. ## How it fits together Install the plugin on a WordPress site and link it to your WPOS account. WPOS reads how the site is built. See [Install](/get-started/install). Run work from the Command Center inside wp-admin: ask for changes, review a plan, and let WPOS execute. See [Command Center](/operate/command-center). Capture how the work should be done as Playbook entries, Skills, and Patterns so it compounds. See [Build memory](/memory/overview). Wire WordPress into the tools you already use through Connectors and MCP. See [Connectors](/connect/connectors). ## Who it is for WPOS is built for WordPress agencies that deliver and maintain client sites, typically on a five figure and up engagement. If you manage one site or fifty, the operating model is the same: connect, operate, remember. # Quickstart Source: https://docs.wpos.ai/get-started/quickstart Connect your first site and run a task in under ten minutes. This guide takes you from nothing to a completed task on a live WordPress site. You need a WordPress site you can install a plugin on, and a WPOS account. If you do not have an account yet, sign up at [app.wpos.ai](https://app.wpos.ai). Install and activate the WPOS plugin on your site. Full instructions, including the manual upload path, are in [Install the plugin](/get-started/install). Open the WPOS panel in wp-admin and connect it to your account. WPOS links the site to your Workspace and reads how it is built. In wp-admin, go to the WPOS menu. This is the [Command Center](/operate/command-center), the surface where you run work. Type what you want in plain language, for example: ```text theme={null} Add a three column "Our services" section to the homepage with icons and short blurbs. ``` WPOS proposes a plan. Review it, approve it, and watch it execute. See [Plan mode](/operate/plan-mode) for how review and approval work. If you want WPOS to do this the same way next time, save the approach as a [Pattern](/memory/patterns) or a [Skill](/memory/skills). This is what makes the next task faster. ## Next steps Learn the full control surface. Make the work compound across sessions. Manage many sites from one place. Bring in Gmail, Slack, and 1000+ other tools. # WordPress, finally operated Source: https://docs.wpos.ai/index WPOS is the operating layer for WordPress. Connect a site, give it memory, and run the work you used to do by hand. WPOS turns WordPress from something you click through into something you operate. You connect a site, WPOS learns how it is built, and from then on it remembers, plans, and executes the work that used to live in your team's heads. This is the operator manual. It covers how to connect sites, run work from the Command Center, build durable memory with Playbook, Skills, and Patterns, and wire WordPress into the tools your agency already uses. Connect your first site and run a task in under ten minutes. The short version of what WPOS is, and what it is not. The control surface inside wp-admin where the work happens. Playbook, Skills, and Patterns. The reason WPOS gets better with use. ## Built for agencies WPOS is built for teams that ship and maintain WordPress sites for clients. It leads with delivery and margin, not credits. If you run multiple sites, see [Workspace](/agencies/workspace) for the cross-site view and [Whitelabel](/agencies/whitelabel) for client-facing setup. WPOS exposes its docs and tools over MCP, so your own Claude or Cursor can operate the product with full knowledge of it. # Build memory Source: https://docs.wpos.ai/memory/overview Playbook, Skills, and Patterns. The reason WPOS gets better with use. Memory is what separates WPOS from a tool you re-explain every session. The enemy is forgetting. Memory is how the work compounds: the second task is faster than the first, and the hundredth is faster than the second. There are three kinds of memory, and they work together. The standing rules. How this site, brand, or client should always be handled. Reusable procedures. A sequence of work WPOS can run on demand. Reusable structures. The shape of a section, page, or component done your way. ## How they fit together Think of it as a flywheel. Connectors bring the outside tools in. Playbook sets the rules. Skills capture the procedures. Patterns capture the structures. Each one makes the others more useful, and every task you run is a chance to add to all three. The rules that should hold no matter what the task is. So the section or page comes out built your way every time. So a repeated sequence of work runs reliably. So the work can pull from and push to the tools you already use. See [Connectors](/connect/connectors). The fastest way to build good memory is to capture it while you work. When WPOS does something the way you want in the [Command Center](/operate/command-center), save it as a Pattern or a Skill right then. # Patterns Source: https://docs.wpos.ai/memory/patterns Reusable structures. The shape of a section, page, or component done your way. A Pattern is a reusable structure: the shape of a section, page, or component as you want it built. Where a [Skill](/memory/skills) is the procedure, a Pattern is the thing the procedure produces, captured so it comes out the same way every time. ## What a Pattern captures * The structure: the parts of the section or page and how they are arranged. * The defaults: spacing, layout, and the elements that should always be present. * The variations: the parts that change per use, and the parts that never do. ## Why Patterns matter Patterns are how you keep a consistent build across many pages and many sites. A "hero section" Pattern means every hero you create starts from the same proven structure, so the work is faster and the result is predictable. This is the difference between an agent that improvises and one that builds the way your team builds. ## Patterns and the rest of memory Brand and convention rules a Pattern must respect. The reusable shape, built to those rules. The procedure that places and fills the Pattern on a real page. When WPOS builds something in the [Command Center](/operate/command-center) that you would want again, save it as a Pattern on the spot. That single action is what makes the next build faster. # Playbook Source: https://docs.wpos.ai/memory/playbook The standing rules for how a site, brand, or client should always be handled. A Playbook is the set of standing rules WPOS follows on a site. It is the place to put the things that should always be true, so you never have to repeat them in the [Command Center](/operate/command-center). ## What belongs in a Playbook * Brand rules: colors, fonts, tone of voice, logo usage. * Conventions: how sections are named, how pages are structured, what an SEO field should contain. * Boundaries: what WPOS should never change without asking. * Client preferences: the specific way this client likes things done. ## Why it matters Every rule in the Playbook is a thing you do not have to say again. It is also a thing a [Guest agent](/operate/guest-agent) and every [Task](/operate/tasks) will follow automatically, which keeps work consistent across your team and across sessions. Playbook is rules. [Skills](/memory/skills) are procedures, and [Patterns](/memory/patterns) are structures. Use Playbook for the things that should hold regardless of the specific task. ## A good Playbook entry A strong entry is specific and testable. Compare: | Weak | Strong | | ------------------ | -------------------------------------------------------------------------------------------------------- | | "Keep it on brand" | "Headings use Inter Tight, body uses the theme default. Primary buttons use #1D2327." | | "Good SEO" | "Every page has a meta title under 60 characters and a meta description between 140 and 160 characters." | Build the Playbook up as you go. The first time you correct WPOS on something that should always be true, that correction is a Playbook entry. # Skills Source: https://docs.wpos.ai/memory/skills Reusable procedures WPOS can run on demand. A Skill is a procedure you have captured so WPOS can run it again reliably. Where a [Playbook](/memory/playbook) holds standing rules, a Skill holds a sequence of work: the steps to get a specific job done your way. ## When to make a Skill Make a Skill when you find yourself describing the same sequence of work more than once. If you have explained "how we set up a new landing page" twice, that is a Skill. ## What a Skill captures * The steps, in order. * The decisions WPOS should make along the way. * The standards each step should meet, drawn from your [Playbook](/memory/playbook). ## Running a Skill Once a Skill exists, you invoke it from the [Command Center](/operate/command-center) and WPOS runs the procedure, showing the plan through [Plan mode](/operate/plan-mode) first. A Skill can drive a whole [Task](/operate/tasks). Skills are procedures. [Patterns](/memory/patterns) are the structures those procedures often produce. A Skill for "build a services section" will usually reach for a Pattern that defines what that section looks like. Keep Skills focused. A Skill that does one job well is easier to reuse than one that tries to do everything. # Command Center Source: https://docs.wpos.ai/operate/command-center The control surface inside wp-admin where you run work on a site. The Command Center is the WPOS surface inside wp-admin. It is where you describe work, review what WPOS intends to do, and watch it execute against the live site. ## What you can do here * Describe changes in plain language and have WPOS carry them out. * Build and edit widgets, sections, and pages. * Edit content, manage plugins, and adjust settings. * Run multi step [Tasks](/operate/tasks) that span several actions. * Review every action before it runs through [Plan mode](/operate/plan-mode). ## How a request flows Say what you want, not how to do it. WPOS works from intent. It uses what it already knows about the site plus your [memory](/memory/overview): Playbook entries, Skills, and Patterns. For anything beyond a trivial edit, WPOS shows a plan first. See [Plan mode](/operate/plan-mode). On approval, WPOS makes the changes on the site and reports back what it did. ## Working with memory The Command Center is also where memory gets built. When WPOS does something the way you want, save it so it happens that way again. This is the difference between a tool you re-explain every time and one that learns your standards. See [Build memory](/memory/overview). The more specific your brand rules and conventions live in [Playbook](/memory/playbook), the less you have to repeat yourself in the Command Center. # Guest agent Source: https://docs.wpos.ai/operate/guest-agent Give a client or teammate a scoped way to run work on a site. The Guest agent is collaboration mode in the [Command Center](/operate/command-center). It lets someone who is not a full operator run scoped work on a single site through WPOS, without handing over your account, your [Workspace](/agencies/workspace), or the rest of your fleet. It is how you bring a client into the loop to request changes themselves, or hand a junior teammate one site to operate, while you stay in control of the standards the work is held to. ## Operator vs guest You. Full access to the site, every Command Center surface, the Playbook, Skills, Connectors, and billing. Sets the rules a guest works inside. A client or teammate working on one site. Can run work through the agent, but cannot change the memory, connectors, or plan that govern the site. ## What a guest can do * Ask the agent to make changes, the same way you would, in plain language. * Watch the work before it lands through [Plan mode](/operate/plan-mode). * Stay inside the guardrails of the site without touching account settings. ## What stays with the operator * The site's [Playbook](/memory/playbook), [Skills](/memory/skills), and [Connectors](/connect/connectors). A guest works inside them, never edits them. * Access to the cross-site [Workspace](/agencies/workspace). The Guest agent is scoped to one site only. * Billing, plan, and site licenses, which always live with the account owner. ## How it stays safe A guest does not get a looser version of WPOS. They get the same one, with a smaller perimeter. Work follows the site's [Playbook](/memory/playbook) and [Patterns](/memory/patterns), so a guest cannot drift from your standards even when they drive the agent themselves. Anything that touches the site can be reviewed in [Plan mode](/operate/plan-mode) before it runs. Cross-site operations stay with full operators. A guest never sees the rest of the fleet. Pair the Guest agent with a strong [Playbook](/memory/playbook) per client. The better the memory, the more you can safely let a guest do without supervision, because the site already knows how it is meant to be operated. Guest agent is a Command Center surface scoped to a single site. To operate many sites at once, use the [Workspace](/agencies/workspace) as a full operator. # Plan mode Source: https://docs.wpos.ai/operate/plan-mode Review what WPOS intends to do before it touches the site. Plan mode puts a review step between intent and action. Before WPOS changes anything beyond a trivial edit, it shows you the plan: what it will do, in what order, and on which parts of the site. ## Why it exists You are operating live client sites. Plan mode means nothing happens by surprise. You see the intended changes, approve them, and only then does WPOS execute. It is the safety rail that makes it comfortable to hand real work to an agent. ## The review loop A clear, ordered list of the actions it will take. Approve as is, ask for changes, or reject. You can tighten the scope before anything runs. It carries out the steps and reports what changed. ## When a plan is shown | Situation | Plan shown | | -------------------------------------------- | ------------------- | | Small content edit | Often runs directly | | Building or restructuring sections and pages | Yes | | Plugin or settings changes | Yes | | Multi step [Tasks](/operate/tasks) | Yes | Plan mode pairs naturally with your [memory](/memory/overview). A well defined [Pattern](/memory/patterns) makes plans shorter and more predictable, because WPOS already knows the shape of the work. # Tasks Source: https://docs.wpos.ai/operate/tasks Run multi step work that spans several actions on a site. A Task is a unit of work that takes more than one action to complete. Instead of issuing each step yourself, you describe the outcome and WPOS sequences the steps, runs them, and reports back. ## Examples of a Task * Stand up a new landing page: create the page, add the sections, set the SEO fields, and link it in the menu. * Roll a brand update across a site: update colors, swap the logo, and adjust the affected sections. * Prepare a site for handoff: tidy settings, check the key pages, and summarize what was done. ## How Tasks run State the end result you want. WPOS breaks the Task into ordered steps and shows them through [Plan mode](/operate/plan-mode). WPOS works through the steps, using your [memory](/memory/overview) so each step matches your standards. You get a summary of what was done across the whole Task. Tasks you repeat are good candidates to capture as a [Skill](/memory/skills), so the same sequence runs reliably every time.