AI Knowledge Center

An AI-powered website is a site whose day-to-day changes are produced by AI coding agents, like Claude Code, Codex, or Gemini, working directly against the site's real codebase, instead of by hand through a CMS. The tools to build and run a website this way already exist; Dispatch makes them manageable for your organization by adding the visibility, context, approvals, and history around every change.

Traditional CMS platforms solved a real problem: structure, permissions, publishing controls, and a safe interface for non-technical teams. AI changes how the work is produced: an agent can create pages, improve content, and make sitewide changes directly against modern website code. Dispatch preserves the CMS's safety net (visibility, approvals, roles, history) without constraining what AI can do. Keep the safety net; remove the dependency on the CMS.

Dispatch connects to your site's repository and hosting platform. Every commit, pull request, and deployment appears in a live activity feed, with AI-authored changes attributed to the agent that made them. Pages are inventoried and scored for answer-engine readiness (AEO), and roles control who can request, approve, and publish. Dispatch governs the work; your coding agent, GitHub, and your hosting platform execute it.

The management layer is the organization-facing system above your coding agent, repository, and hosting platform. It holds what those tools don't: who requested each change, what brand context governed it, who approved it, what it did to your pages, and how to undo it. Dispatch gives marketing teams the visibility, context, approvals and control they need to operate a website built and managed by AI.

It is safe when the same controls a CMS provided still exist: every change previewed before it ships, approvals owned by named roles, AI-authored work attributed to the agent that made it, and a restore path back to any prior version. Git-based sites make each of these stronger than most CMS workflows. What makes AI unsafe is not the agent; it is running one with no management layer around it.

A change starts as intent, a request written in plain language. Dispatch turns it into a scoped agent handoff, the agent produces the work on a branch, and the deploy platform builds a preview. The approver reviews the preview and either approves or sends it back with a revision note, which regenerates the agent's instructions on the same request. Roles are enforced so editors cannot approve their own work, and every decision is recorded next to the change it authorized.

Coding agents sign their commits with attribution trailers: Claude Code, Codex, and Gemini all identify themselves in the commit message. Dispatch reads those trailers and badges each change in the activity feed with the agent that made it, alongside the human who requested and approved it. Attribution is what turns AI website work from a black box into an auditable record.

You restore it in one click. Dispatch lists every prior version of your site, and an owner or admin can restore to any of them. A new commit puts the site back exactly as it was; nothing is deleted, and both the mistake and the recovery stay on the record. Deploy previews catch most problems before production; for anything that slips through, the activity feed shows exactly which change caused it and the Versions panel undoes it.

Not if the migration preserves what search engines care about: URLs and redirects, titles and descriptions, canonical tags, sitemaps, and structured data. A code-first site usually improves technical SEO because you control every tag directly instead of working through CMS templates. Dispatch scores every page after each sync, so regressions in those signals are visible instead of silent.

An AEO score measures how ready a page is to be cited by AI answer engines like ChatGPT, Claude, Perplexity, and Google AI Overviews. Dispatch scores every page from 0 to 100 across eight weighted checks: title, meta description, H1, canonical URL, structured data (weighted heaviest), social tags, indexability, and HTTP status. It rolls the results up so you can see your site's biggest issues at a glance.

You need less routine development, not zero engineering judgment. Agents handle the production work: new pages, copy changes, structured data, and sitewide updates. A technical partner matters for initial setup, review standards, and the occasional complex change. Many teams run day-to-day operations from the management layer and involve developers the way they involve legal: for the decisions that warrant it.

Webflow is an excellent platform, and its Enterprise governance (staging, permissions, publishing controls) is the model of the safety net organizations need. Dispatch serves teams that want that same safety net without the CMS as the production environment: agents work on a real codebase, and Dispatch supplies the governance above it. Move beyond the limitations of your CMS without giving up the visibility, control and governance your team depends on.

Inventory your current site, rebuild it on a modern framework with design and content preserved, wire redirects and SEO signals, load your organizational context, and connect governance before the first AI change ships. Teams do this in-house or with a migration partner; Dispatch's Expert Onboarding includes a partner-led path where the site arrives already governed. Verify fidelity page by page before cutover.

No. The site is rebuilt to match, and every page is checked pixel for pixel against your live site before cutover. Where a page differs, the difference is fixed or shown to you and agreed before launch. Design updates, if you want them, are scoped separately so the migration itself stays a like-for-like move.

It moves with the site. Pages, posts, images, and metadata are exported from your current platform and rebuilt in the new content model: git-based content for most business sites, a headless CMS or database where the site needs one. The blog keeps its URLs, and your team keeps editing through plain-language requests that an agent carries out.

Yes. Forms are rebuilt and wired to wherever submissions go today, every old URL is mapped to its new home with redirects, and analytics and conversion tracking are reinstalled and verified before cutover. Sitemaps, structured data, and search console are part of the launch checklist.

Most business sites migrate in two to four weeks from kickoff to cutover. Larger sites, heavy integrations, or app-like functionality take longer. The assessment says how long before any build work starts, and the quote is fixed to that scope.

The Dispatch migration team with partner agencies. You get named contacts once the engagement is agreed, one point of contact through cutover, and the same team stays reachable after handoff.

No. Every migration lands connected to Dispatch on the free plan, with the repository and host connected, pages scored, and approvals on, because that is how we hand a site back governed. Whether you keep operating it through Dispatch is up to you. The code is yours, in your repository, with no lock-in.

Connect your site's GitHub repository and your hosting platform (Vercel, Netlify, or Cloudflare) from the Websites section. Dispatch syncs your pages, scores each one for AEO readiness, and starts recording every commit, pull request, and deployment in the activity feed with attribution. From there, add your brand context and set roles so approvals are enforced from the first change.

Dispatch is agent- and host-agnostic. It works with any coding agent (Claude Code, Codex, Gemini, and whatever comes next), GitHub for version control, and Vercel, Netlify, or Cloudflare for hosting and deploys. Approval alerts reach your team in Slack and by email. Because Dispatch sits above the execution tools rather than replacing them, every improvement they ship makes your setup better.

Yes. Dispatch re-crawls every governed page on a schedule you set per site, keeps a score history for each page, and opens a finding when a check that used to pass now fails, a score drops, a page stops answering, or a page falls out of the sitemap. Each finding carries a drafted change request an agent can execute, so the work arrives already written. Findings close on their own when the check passes again. This is in production today.

Only when you tell it to, and only for low-risk requests. Each site has a policy that lets low-risk requests go live without waiting for a reviewer. The approval is recorded with the policy as the actor, the activity feed logs it, and restore is one click away. Standard-risk requests still wait for an approver, and high-risk requests need two distinct approvers, enforced in the database rather than in the interface.

Not all of them, and not on day one. Dispatch governs websites that live as code in a GitHub repository and deploy through Vercel, Netlify, or Cloudflare. A client whose site fits comfortably inside Webflow, WordPress, Wix, Squarespace, Framer, or HubSpot CMS can stay there; the agency's judgment about which clients have hit the platform ceiling is the point. When a client is ready to move, the migration itself runs as governed work, so the record starts with the first commit rather than after cutover.

The client. A client site belongs in the client's own Dispatch organization from the start, with agency staff invited as members, so ownership never has to be transferred out of the agency's account. At handoff the client holds the owner role, every repository, host, and Slack token is rotated to one the client controls, and the agency keeps whatever role the engagement calls for: admin to keep operating, editor to keep filing requests, or a standing escalation contact with no seat at all. The client's team lands on the Inbox and approves; the record, restore, and the alerts stay with the site.

Yes. Each client site connects with its own repository, host, policies, approvers, and context, and every governed site reports into one Inbox where approvals, findings, and slipped pages are grouped by site. Risk tiers, crawl cadence, and low-risk auto-approve are set per site, so a cautious client and a fast-moving one can run side by side. Plans scale by the number of governed websites; agency plans with client portfolios are arranged through sales.

Yes. Every change request gets its own branch and preview, so people and agents work in parallel without overwriting each other. Roles decide who can approve, high-risk requests need two distinct approvers, and every action lands in one activity feed with the person or agent attributed. The Inbox shows the whole team what needs them, and Slack alerts carry the draft when someone is needed.

The request. Every job an agent gets from Dispatch is scoped to a page, a risk tier, and your approved context, and the agent works on a branch, never on production. Your team reviews the preview before anything ships, a second approver is required for high-risk work, and one click restores the previous version if something still slips through.

An AI Operating System is a centralized platform that helps organizations organize, share, govern, and improve all of their AI knowledge, prompts, agents, workflows, documentation, and institutional expertise. Instead of AI living inside individual chat conversations, it becomes a shared organizational asset that teams can continuously build upon.

As AI adoption grows, organizations often accumulate hundreds of prompts, workflows, and AI assets across different teams. An AI Operating System provides a single source of truth that helps employees discover, reuse, improve, and govern AI resources rather than recreating them.

No. ChatGPT is an AI application. An AI Operating System organizes how your entire organization uses AI applications such as ChatGPT, Claude, Gemini, Microsoft Copilot, custom GPTs, and AI agents.

An AI Operating System helps solve:

• Duplicate prompts
• Lost AI knowledge
• Inconsistent AI outputs
• Lack of governance
• Poor collaboration
• Knowledge loss when employees leave
• AI adoption challenges
• Difficulty scaling AI across departments

Organizations with multiple employees using AI, especially marketing teams, sales teams, operations, HR, customer service, leadership, and agencies, benefit from an AI Operating System because it centralizes knowledge and improves consistency.

Organizational AI is the practice of applying artificial intelligence across an entire business rather than through isolated individual usage. It emphasizes shared knowledge, collaboration, governance, and long-term organizational learning.

A traditional knowledge base stores information. An AI Operating System stores information while also organizing prompts, AI agents, workflows, automations, governance policies, brand context, reusable templates, and collaboration around AI.

Assistant memory is personal and tool-specific: what ChatGPT remembers for one teammate is invisible to a colleague on Claude, Gemini, or Grok, and it disappears with the account. A company needs shared, durable knowledge it owns. Dispatch acts as that knowledge base: approved brand voice, context, prompts, and workflows stored once, governed by approvals, and served to every AI tool your team uses through a governed connection.

No. Small businesses often experience AI chaos much sooner because fewer people are responsible for many different roles. An AI Operating System helps organizations of all sizes organize their AI knowledge from the beginning.

Every AI Operating System should include:

• Prompt library
• AI agent library
• Workflow documentation
• Brand guidelines
• AI governance
• Search
• User permissions
• Context documentation
• Team collaboration

The greatest benefit is turning individual AI experimentation into institutional knowledge that every employee can use.

AI governance is the framework an organization uses to ensure AI is deployed consistently, responsibly, securely, and in alignment with company standards and business goals.

Without governance, employees may unknowingly create inconsistent messaging, expose sensitive information, duplicate work, or rely on outdated prompts and workflows.

No. Good governance accelerates innovation by giving employees trusted resources and proven starting points rather than forcing them to reinvent everything.

A governance policy should define approved AI tools, security guidelines, prompt standards, content review processes, brand requirements, and ownership of AI assets.

AI governance is most effective when shared between executive leadership, operations, IT, legal, and department leaders.

Yes. Centralized prompts, approved templates, and shared brand context help ensure AI-generated content reflects a consistent voice across the organization.

No. Governance also includes collaboration, knowledge sharing, quality control, version management, and continuous improvement.

Organizations should establish governance as soon as multiple employees begin using AI regularly.

Prompt management is the process of organizing, documenting, categorizing, improving, and sharing prompts across an organization.

Chat conversations are difficult to search, impossible to govern across teams, and easily lost over time.

A quality prompt library includes categories, tags, ownership, search, descriptions, and examples — so anyone can find and reuse the best work.

Yes — prompts should be treated as living assets. Dispatch tracks each prompt’s current approved version and when it was last updated; deeper revision history is on our roadmap.

Prompts should be categorized by department, function, use case, AI model, business objective, and audience.

Absolutely. High-performing prompts often represent significant organizational knowledge and competitive advantage.

Every employee should be encouraged to contribute, while designated owners review and approve prompts for broader organizational use.

Without centralized prompt management, employees often have no visibility into prompts already created by others.

Teams collaborate by sharing prompts, workflows, AI agents, documentation, and lessons learned through a centralized platform.

Most AI knowledge remains inside private conversations, making it inaccessible to the rest of the organization.

Yes. Shared AI assets help marketing, sales, HR, operations, and leadership build upon one another's work rather than starting from scratch.

Yes. Search is one of the most valuable features of an AI Operating System because employees need to quickly discover existing knowledge.

Institutional AI memory refers to preserving an organization's AI knowledge beyond individual employees.

Without centralized documentation, much of their AI expertise may leave with them.

Yes. Documenting AI workflows makes them repeatable, measurable, and easier to improve over time.

Absolutely. New employees can begin using proven prompts, workflows, and AI resources immediately.

Successful organizations focus on people, processes, governance, and knowledge sharing, not just technology.

Common barriers include lack of training, poor organization, inconsistent usage, security concerns, and limited visibility into existing AI efforts.

Yes. Each department has unique use cases, but all should operate within a shared organizational framework.

Scaling AI requires centralized knowledge management, governance, standardized workflows, and ongoing collaboration.

In most organizations, AI enhances employees by automating repetitive work and enabling people to focus on higher-value activities.

Organizations can measure adoption through time savings, reuse of AI assets, workflow automation, knowledge sharing, and business outcomes.

AI maturity describes how effectively an organization has integrated AI into its people, processes, governance, and operations.

The next stage is organizational optimization, continuously improving and expanding shared AI knowledge.

Dispatch is an AI Operating System that helps organizations organize, govern, search, share, and continuously improve their AI knowledge.

The Inbox is the page in Dispatch where work waits for you. It holds change requests with a preview ready for your approval, findings from the scheduled crawl of every governed page (each with a drafted fix), pages whose score slipped since the last crawl, and the work you have in flight. Approvers land on it after login. Dispatch finds the work, drafts the fix, and puts it in front of the right person; you keep the approve moment.

Dispatch is designed to manage prompts alongside AI agents, workflows, documentation, governance, brand assets, context, and organizational knowledge.

Dispatch is built for organizations that want to scale AI across teams while maintaining consistency, collaboration, and governance.

Yes. Marketing, sales, HR, operations, executive leadership, customer service, and other teams can all organize and share AI resources within a single platform.

No. Dispatch complements AI tools by organizing how organizations use them.

Dispatch makes AI knowledge searchable, reusable, governed, and accessible across departments.

Centralization reduces duplication, preserves institutional knowledge, improves consistency, and accelerates AI adoption.

Begin by documenting your existing prompts, workflows, AI tools, and best practices. From there, create a centralized AI Operating System that allows every employee to build upon the organization's collective knowledge.

No. Your website code, content, and history live in your repository and host, not in Dispatch; turn Dispatch off and the site keeps running exactly as it is. Everything you create inside Dispatch, prompts, context, agents, workflows, documents, and the full request and approval record, exports from Settings as JSON and Markdown at any time.