FAQs

Frequently asked questions

What Garchi CMS is, what it manages, and how your team, your application and your AI tools work with it.

Garchi CMS is a headless content management system for websites and applications. It stores your structured business content separately from your frontend and makes that content available through the Garchi dashboard, a REST API, and MCP for compatible AI agents.
No. Garchi manages the content layer. You build the frontend with your own framework, your developers or an AI coding tool, and connect it to Garchi to fetch the content it needs.
Websites and applications can be built quickly, but the content inside them keeps changing: headlines, page copy, articles, product information, categories, images, documentation and translations. Without a content layer, each of those routine changes means editing application source code or asking an AI coding agent to modify the frontend again. Garchi gives that content a structured home outside your code. Presentation and application changes — new components, new layouts, new functionality — still happen in your codebase.
Startup founders, SMEs, product teams, developers, freelancers and software agencies who want business content managed separately from frontend code. It is a good fit for teams building with AI coding tools, where the generated application should not also become the editing interface for routine content.
A headless CMS stores and manages content separately from the frontend that displays it. You fetch the same content through an API and reuse it across websites, mobile apps, portals and other digital experiences.
Yes, provided your frontend can fetch content over HTTP. You create a space in Garchi, model the content you want to manage there, and update the relevant parts of your frontend to read that content from the Garchi API or SDK. How much work that is depends on how your current content is stored — it is an integration, not a zero-effort switch.
Garchi exposes a REST API, so any framework that can make an HTTP request can consume it. Node.js and PHP SDKs are available if you prefer them, and starter kits are provided for Next.js, Nuxt, SvelteKit and Laravel.
Headless pages built from reusable sections and section templates, structured data items such as articles or products, categories, key-value metadata on items, uploaded assets, and language-specific content for multilingual sites.
Data items are flexible structured content records — a blog article, a product, an event, or anything else itemable. They start from a base schema and are extended with metadata such as authors, ratings, tags or any other key-value attributes your content needs, without changing your own application's database schema.
Yes. You add the languages you need to a space, and pages and sections can hold language-specific content that you manage in one place. The number of additional languages available depends on your plan.
In most cases, no. Garchi is intended for business content that needs to be managed and published. Your application can keep using its own database for users, transactions, permissions and application-specific state.
Yes. You can invite collaborators to a space and set what each of them is allowed to do in that space, so content work does not have to go through your deployment pipeline.
MCP, the Model Context Protocol, is an open protocol that lets compatible AI tools connect to external systems. Garchi runs an MCP server, so an authorised AI tool can work with the structured content in your Garchi account.
Yes, through Garchi's MCP server. Once you connect a compatible AI client, it can work with the content available to the connected Garchi account — listing spaces, creating and updating pages and sections, managing data items, categories and metadata, and uploading assets. It works against your real content model rather than inferring content from your frontend source code.
Any AI client that supports the Model Context Protocol and can connect to a hosted MCP server. Garchi is not affiliated with or endorsed by those tools' vendors — compatibility simply means the client speaks MCP.
Yes. The Garchi dashboard is the primary way people work with content. MCP is an additional way in, not a requirement.
Through the Garchi API. You create an API token from your dashboard and send it in the Authorization header as a Bearer token when making authenticated requests. API requests are subject to technical rate limits and to the bandwidth included with your plan — see the API documentation for current limits.
Sign up for the free Sandbox plan, create a space, add some content and generate an API token. From there you can call the API directly, use one of the starter kits, or connect a compatible AI tool through MCP.

Still have a question? Get in touch or read the documentation.