- Brand
- Garchi CMS
- Category
- Content & Design
- Primary Subcategory
- CMS & Website Content Management
Integration details
Description
Garchi CMS is a headless SaaS CMS used to manage content across platforms such as websites and apps. With this app, you could connect to Garchi CMS using the MCP server and manage your platform's content levaraging the power of Chat GPT. Example prompts - Create a blog article on best principles of Gen AI using Garchi CMS - Create a landing page using the provided brief in Garchi CMS.
- Integration type
- Plugin
- Verification status
- Not applicable
- Platform
- ChatGPT
- Primary Subcategory
- CMS & Website Content Management
- Secondary Subcategories
- None listed
- Brand
- Garchi CMS
- Access
- Account required
- First tracked
- 2026-08-24
- Tool count
- 43
- Market
- US
The Primary Subcategory used for this profile’s headline score.
Other Subcategories where the Integration is listed.
Your score is coming
ChatGPT now suggests Plugins on its own when they match a user's request.Your Plugin Discovery Score measures how often yours appears, and it will show here as soon as it’s ready.
What discovery looks like

Get alerts for Garchi CMS
Get updates when Garchi CMS’s Discoverability Score or category rank changes.
Competing in ChatGPT CMS & Website Content Management
View Category43 tools agents can invoke
This tool is used to add a new language to the space for translation purposes of the pages.
add-language-to-space-tool
This tool changes the rank/order of a section within a nested section on a page. For example if a page has 3 sections in the following order under Featured Section: - Section A (order 0) - Section B (order 1) - Section C (order 2) If we want to move Section C to be between Section A and Section B, we would change its order to 1. The page is put into draft; a previously published version, if any, keeps being served until the draft is published.
change-nested-section-rank-tool
This tool changes the rank/order of a section within a page. For example if a page has 3 sections in the following order: - Section A (order 0) - Section B (order 1) - Section C (order 2) If we want to move Section C to be between Section A and Section B, we would change its order to 1. The page is put into draft; a previously published version, if any, keeps being served until the draft is published.
change-section-rank-tool
Creates a new category in a space. Categories group data items, such as blog articles or products; every data item belongs to at least one category, given by the category's id. The new category is placed after the space's existing categories. Creating a category changes no data item and no published content. Calling it twice creates two categories. Creating counts against the plan's category allowance. Returns the created category with its id.
create-category-tool
This tool creates a new data item in the Garchi CMS. Data items can be blog articles, products, documents or any itemable content type defined in the CMS. Each data item has one or more categories associated with it, so valid category IDs are required. The item is created as a draft (`published: false`) and stays unpublished until it is published from the Garchi dashboard or, if `scheduled_for_datetime` is provided, until that time. An item can go inside another data item of the same space through `parent_id`, such as a variant inside its product or a lesson inside its course; it is added after the items already inside that parent. This tool does not set the item's image. The item's main image is set separately, after the item is created, using the returned item id. **This tool is for data items only, not for headless web pages or sections.**
create-data-item-tool
Creates a new item type in a space: the blueprint for one kind of data item, such as Product, Event or Changelog entry, with a fixed set of fields that every item of that kind shares. A type is created either from a ready-made `template` or from a `name` and a list of `fields`. Each field has a `label` and a `kind`; its `key` is made from the label when none is given, and is the metadata key the field's values are stored under. A choice field needs `choices`, and a group field needs `group_fields`. Creating a type changes no data item and no published content. Items are linked to the type afterwards through their `item_type_id`, and their values are stored as ordinary metadata entries, so the content API returns them like any other metadata. A name already used by a type in the space is refused, and the error gives that type's id. Returns the created type with, for each field, the metadata `key`, `meta_type` and `value_format` to fill it with.
create-item-type-tool
Every data item can have key value pair of metadata associated with it to extend its information. These are key value pairs where the key is a string and the value is stored as a string in all cases. Type determines how the string should be interpreted: - string, url, email, color, icon_lucid, icon_hero: plain string value - numeric: numeric string (e.g. "42" or "19.99") - date: date string (e.g. "2026-07-08") - array: JSON-stringified array (e.g. "[\"a\",\"b\"]") - object: JSON-stringified object (e.g. "{\"author\":\"Ada\"}") Example: author as key with value "author_name" and type "string" for a blog article item. This tool helps in creating metadata entries for data item. Each item can have multiple metadata entries. Adding metadata sets the data item to draft (`published: false`); it stays unpublished until it is published from the Garchi dashboard or its pending scheduled publication, if any, runs.
create-meta-for-item-tool
This tool adds a new nested section to a section on an existing headless web page. The section is based on a specified section template; this tool does not accept prop values. This tool nests the new section within a parent section, which is useful for creating complex page structures. The page is put into draft; a previously published version, if any, keeps being served until the draft is published.
create-nested-section-tool
This tool creates a new headless web page in a specified space. This tool doesn't create page content, rather it creates the page shell with metadata like title, description, path, and optional JSON-LD for SEO. Sections are added to the page separately. The path and the title must each be unique within the space.
create-page-tool
This tool creates a new section template (or, if the name already exists in the space, replaces that template's props and description; published pages using it then become drafts) that can be used to add sections to headless web pages. Section templates define the structure and content placeholders for sections. Section templates can include `props`, where each prop must follow this shape: - key: lowercase snake_case only (letters, numbers, underscores) - type: one of text, longtext, media, richtext, select, icon_lucid, icon_hero, date - allowed_values: optional string, but required when type is select (comma-separated, e.g. left,right) For icon prop types (`icon_lucid`, `icon_hero`), the value should be a valid icon name from the corresponding icon library.
create-section-template-tool
This tool adds a new section on an existing headless web page. The section is based on a specified section template; this tool does not accept prop values. The page is put into draft; a previously published version, if any, keeps being served until the draft is published. With this tool you can put section templates as sections on your headless page.
create-section-tool
Creates a social post as a DRAFT in Garchi. Nothing is published by this tool. **A post needs at least one connected channel.** `channel_ids` takes the `channel_id` of every connected channel the post goes to (one or more). A space with no connected channel cannot have posts; LinkedIn, Instagram or Facebook are connected from Social Publishing in the Garchi dashboard. This tool refuses a post without a valid connected channel, and creates nothing when it refuses. A person approves the post in the Garchi dashboard and then schedules it or publishes it immediately; Garchi then publishes it to the channels. **Media must already exist in Garchi. This tool never uploads anything.** - `space_asset` — an asset from the space's library. Give the asset's id. - `data_item_image` — a data item's feature image. Give the data item's id. The feature image filename is captured now. If someone later changes that data item's feature image, this post keeps the image that was chosen, which is deliberate. If you set `source_type` and `source_id`, the post records which page or data item it came from. The post is then independent: editing that page or item later does not change this post.
create-social-post-tool
Deletes one metadata entry from a data item. Other metadata entries on the item are not changed. The entry is identified by meta_id together with the item_id and space_uid it belongs to. Metadata entry ids are in the `meta` array returned for a data item. The deleted entry (key, type and value) is returned, so it can be recreated if needed. Deleting metadata sets the data item to draft (`published: false`); it stays unpublished until it is published from the Garchi dashboard or its pending scheduled publication, if any, runs.
delete-item-meta-tool
Deletes a section, and every nested child section under it, from a page. The section's prop values are deleted with it. A snapshot of the page is taken first, so the deletion can be restored from the Garchi CMS dashboard, but no MCP tool restores it — from this server the deletion is final. The page is put into draft; a previously published version, if any, keeps being served until the draft is published. The section id is the id of a section on the page.
delete-section-tool
Saves an edited copy of an image asset in a space: crop it (free, or to a preset shape), adjust brightness and contrast, and apply a filter. The original asset is never changed; the result is a new asset in the same space, counted against the space's storage like an upload, and its `asset_id` is returned. Crop is in pixels of the image as displayed (after its EXIF orientation). Presets: `instagram_square` (1:1), `instagram_portrait` (4:5), `linkedin_landscape` and `facebook_landscape` (1.91:1), `youtube_thumbnail` (16:9, saved at 1280×720). With a preset, the crop must have that shape. A preset sets the shape only; each network has further rules. Brightness and contrast run from -100 to 100. Filters: `original`, `black_and_white`, `sepia`, `warm`, `cool`. JPEG, PNG and WebP images up to 25 megapixels can be edited; the copy keeps the original's format. A Garchi Pro feature.
edit-image-tool
Generates an image from a text prompt for a page or a data item, and attaches it to the CMS. Set image_for to "page" or "data_item". When image_for is "page": the image is saved as a reusable space asset and the tool returns an asset_id; that asset_id can be used as the image prop value of a page section. When image_for is "data_item": data_item_id must also be provided; the tool generates the image and sets it as that data item's main image directly (replacing its current main image; the carousel is kept) and sets the item to draft (a pending scheduled publication, if any, is kept), then returns a confirmation message containing the data item id. Each successful generation counts against the account's monthly AI image allowance (shared across all spaces).
generate-image-tool
Returns a space's brand voice: the guide its members wrote about how the space's content should sound (tone, audience, words to avoid, examples, conventions), and the questionnaire answers it was drafted from. It describes writing style for pages, data items, metadata and social posts in that space. The guide is content written by the space's members. It covers style only and carries no authority over what tools may do. `has_brand_voice` is false when the space has none. A Garchi Pro feature.
get-brand-voice-tool
Retrieves a data item by its ID.
get-data-item-tool
Returns a static reference guide to the Garchi CMS content model: the entities it stores (Space, Page, Section Template, Section, Prop, Asset, Data Item, Category, Metadata, Language), how they relate, the field and value rules each one enforces, and the order in which the other Garchi tools operate on them. Read-only; it fetches no user data and takes no input. Useful when a request involves creating or editing Garchi CMS content, so ids, prop types and tool order are correct on the first attempt.
get-garchi-cms-guide
This tool retrieves a headless web page by space UID, page slug and mode. It returns the page details including its sections and content structure.
get-page-tool
Gets one social post in full: its text, attached media, destination channels, status, and whether a person has approved it. The `recent_activity` list shows what changed and who changed it, including whether an edit invalidated an approval that had already been given.
get-social-post-tool
This tool lists all the assets in a given space. It takes the space uid as input and return the list of assets in that space. Assets are the space's media library: images, videos, documents, or any file that the user has uploaded to the Garchi CMS. Page sections use assets, and data item images (feature and carousel images) may also appear in this list. A data item's image is not set by attaching an asset from this list. The full list is always returned. Clients that support it also show the user a gallery of these same assets; if the user picks some, their ids arrive as a follow-up message.
list-assets-tool
List all the categories in a given space. Categories are associated only with data items.
list-categories-tool
Lists the content calendar of a space between two dates: social posts that are scheduled, waiting for approval with a planned time, published or failed; data items scheduled to publish; and data items and pages by the date they were last updated. Read-only: scheduling and editing happen elsewhere. Each entry has a `source` (social, data_item or page), a `kind` (scheduled, planned, published, failed or updated), a title, a status and a time. Social and update times are ISO 8601 in UTC. A data item's scheduled time is `floating`: it is the wall-clock time exactly as it was entered, with no time zone. Pages cannot be scheduled and have no stored publish date. A Garchi Pro feature. The range can be up to 93 days.
list-content-calendar-tool
This tool lists all data items in the Garchi CMS for a specified space. Data items can be blog articles, products, documents or any itemable content type defined in the CMS. Each item's `parent_id` is the id of the data item it is inside (a variant inside its product, a lesson inside its course), or null when it is on its own, so the list shows how items are nested.
list-data-items-tool
This tool lists metadata keys and types for items in a space. This is useful to keep it consistent while creating item's metadata. Data items have limitation in terms of fields, thus metadata is used to extend them.
list-item-meta-tool
Lists the item types of a space, with their fields. An item type is the blueprint for one kind of data item, such as Product, Event, Team member or Changelog entry: a name and a fixed set of fields, so every item of that kind carries the same structured values. A data item is linked to a type by its `item_type_id`. The type's values on an item are ordinary metadata entries: for each field, `key` is the metadata key, `meta_type` is the metadata type, and `value_format` says how the value string is written. Fields with `required: true` are expected on every item of the type. `items_count` is how many data items already use each type. `templates` lists the ready-made types a new type can be created from, by template key. `data` is empty when the space has no item types yet.
list-item-types-tool
This tool is useful to list all added languages in a space. Languages can be added to a space. They are used to manage the page content in different languages within the same space. All spaces use en-US as the default language.
list-language-tool
It list all the headless pages in the workspace. This list can be used to understand what pages exist in the CMS and get information about them.
list-pages-tool
This tool lists all the section templates available in a given space. Section templates define reusable sections that can be added to headless web pages. They can be mapped to React or Vue components to render dynamic content. Each section template has props. Props define the configurable fields for the section template.
list-section-template-tool
Lists the social media channels a space has connected, such as a LinkedIn account. Use this before creating a social post, to get the `channel_id` values a post can be sent to. A channel with `usable: false` cannot be published to until the user reconnects it in the Garchi dashboard. Connecting and disconnecting channels is done by a person in the dashboard, not through this server.
list-social-channels-tool
Lists social posts in a space, newest first, optionally filtered by status. Statuses: `draft` (being written), `pending_approval` (waiting for a person), `approved` (signed off), `scheduled`, `publishing`, `published`, `failed`, `cancelled`. Returns a summary per post. A single post can be retrieved in full, with its text, media and channels.
list-social-posts-tool
List all spaces owned by the authenticated user. Use this tool when you need to: - discover available spaces/workspaces before creating or querying content, - get a valid `space_uid` for other space-scoped tools, - inspect high-level space metadata and content counts. Returns an object with a `data` array of space objects, each with: - `space_uid` - `name` - `agent_description` - `pages_count` - `items_count` - `categories_count` - `section_templates_count` - `assets_count` - `front_end_url` (optional, if available) Search keywords: list spaces, list workspaces, get spaces, get space uid, available spaces, user spaces.
list-space-tool
Marks a draft as ready and hands it to a person for approval; the post moves to `pending_approval`. A person approves the post in the Garchi dashboard and then schedules it or publishes it immediately; Garchi then publishes it to the channels. This tool does not approve, schedule, or publish the post. Before this succeeds the post must be in draft or pending_approval status and have at least one channel, and its content must satisfy each destination network's rules — for example LinkedIn allows 3000 characters and a single image. If validation fails, the error names the channel and the reason.
request-social-post-approval-tool
Saves a space's brand voice. `guide` replaces the whole guide (Markdown, up to 20,000 characters; an empty string clears it). `answers` replaces the questionnaire answers. Whichever of the two is left out keeps its current value. Answer keys: brand_name, business, audience, values, avoid, examples, social, conventions (text); tones (up to 6 of: Friendly, Professional, Confident, Warm, Playful, Calm, Bold, Inspiring, Authoritative, Witty, Empathetic, Straightforward); formality (very_formal, formal, balanced, conversational, very_conversational); spelling (british, american, other). A Garchi Pro feature.
update-brand-voice-tool
Renames an existing category of a space. The new name replaces the old one everywhere the category is shown, including on every data item in it. The category keeps its id, its position and its data items, and those items keep their published state. Renaming changes nothing else. Returns the renamed category.
update-category-tool
Updates a single existing data item (such as a blog article, product, or document) within a space. Provide the data item id and the space_uid it belongs to, along with only the fields you want to change — any field you omit is left unchanged. When categories are provided, they replace the item's existing categories, so the full intended list must be provided. The item is set to draft (published: false) and stays unpublished until it is published from the Garchi dashboard or until its scheduled publication time, if `scheduled_for_datetime` is provided or a schedule is already pending. This tool updates one item per call and is only for data items, not for headless web pages or sections. This tool does not change the item's image.
update-data-item-tool
Updates one existing metadata entry on a data item: its key, its type, its value, or any combination. Fields left out keep their current value; at least one of key, type or value is required. The entry is identified by meta_id together with the item_id and space_uid it belongs to. Metadata entry ids and their current values are in the `meta` array returned for a data item. The value is always stored as a string and is checked against the entry's type (the new type when type is also given): - string, icon_lucid, icon_hero: plain string - url, email: a valid URL or email address - color: hex colour (e.g. "#1a2b3c") - numeric: numeric string (e.g. "42" or "19.99") - date: date string (e.g. "2026-07-08") - array: JSON-stringified array (e.g. "[\"a\",\"b\"]") - object: JSON-stringified object (e.g. "{\"author\":\"Ada\"}") Changing metadata sets the data item to draft (`published: false`); it stays unpublished until it is published from the Garchi dashboard or its pending scheduled publication, if any, runs.
update-item-meta-tool
This tool updates an existing headless web page with new metadata like title, description, path, and optional JSON-LD for SEO. Only the fields provided in the input will be updated, allowing for partial updates to the page. If the page is published, its current sections and page fields (title, path, description, JSON-LD, agent description) are saved as a version first and the page is put into draft, so the live site keeps serving the last published state, including its previous path, until a person publishes.
update-page-tool
This tool updates an existing prop template for a section template. You can update any one field (`key`, `type`, or `allowed_values`), any combination of two fields, or all three fields together. Supported fields: - key: lowercase snake_case only (letters, numbers, underscores), e.g. background_image - type: one of text, longtext, media, richtext, select, icon_lucid, icon_hero, date - allowed_values: comma-separated string, e.g. left,right,center Consistency rules: - If `type` is set to `select`, `allowed_values` is required in the same request. - If `type` is not `select`, `allowed_values` is optional. Example: updating a prop template key from background_image to hero_image and type from text to media. Scope of the change: a prop template belongs to a section template that may be reused across many pages. Changing a `key` or `type` therefore changes every section built from that template, and can leave existing prop values invalid for the new type. Published pages using the template keep serving their last published content and become drafts, so the change goes live only when a person publishes them.
update-prop-template-tool
Updates an existing social post. Only the fields you provide are changed; omitted fields are left alone. **Important — approval is not preserved across edits.** If a person has already approved (or scheduled) this post and an edit changes the text, the media, or the channels, that approval is automatically withdrawn and the post goes back to `pending_approval`; it must be approved again before it can publish. The same applies to edits made in the dashboard. An edit that leaves the publishable content unchanged keeps the approval. The response reports when approval was withdrawn. `media` REPLACES all existing attachments — send the full list you want, or omit it to leave the current media untouched. Media must already exist in Garchi; nothing is uploaded. Only posts in draft, pending_approval, approved or scheduled status can be edited; a post that is publishing, published, failed or cancelled cannot.
update-social-post-tool
This tool uploads an asset (image, PDF, document, text or CSV file) into a space. For images, PDFs and documents, the tool needs only space_uid, file_name and file_type: it opens an upload window for the user to pick the file from their device, and also returns a secure signed upload_url for hosts that cannot render that window. File bytes for these types are supplied by the user through the upload window or signed URL, not through tool arguments. This tool does not generate images. A small text/plain or text/csv file may be written directly by passing its plain text as file_raw_content. Uploaded assets are added to the space's media library and can be used in page sections; a data item's image is not set by uploading an asset.
upload-asset-tool
This tool creates or updates a section's content on the page. Upsert merges by prop id: provided props are overwritten or added, omitted props are left untouched. Each section is associated with a section template that defines prop ids and types. For each `props` item, send: - `id` (required): prop template id - `value` (required for non-media props): text/select/date/etc value - `asset_id` (required for media props): asset id from the space assets list Consistency rules: - If prop type is `media`, `asset_id` is required and `value` is ignored. - If prop type is not `media`, `value` is required and `asset_id` must be omitted. - If prop type is `select`, `value` must be one of that prop's allowed values. The page is put into draft; a previously published version, if any, keeps being served until the draft is published.
upsert-section-content-tool
Garchi CMS ChatGPT Plugin FAQ
How the directory, categories and Discoverability Score work.
Read the methodologyHow do I improve Garchi CMS's ChatGPT Plugin discoverability?
The levers are the listing surface agents actually read: names, descriptions, keywords, tool metadata, and registry health. Which lever matters depends on where discovery breaks, which is what continuous measurement shows.
What are Garchi CMS alternatives on ChatGPT?
As of 2026-11-02, Garchi CMS competes with Adobe Experience Manager, AI Engine for WordPress, Atmoscreen, BrightSite, CMDPress, Contentful, Digital Vibes Echo, Docmino and 17 more in ChatGPT CMS & Website Content Management, ranked by public Discoverability Score.
Where is this profile measured?
This profile uses the geography attached to the latest public registry snapshot: US. Locale tags are intentionally omitted.