Summary
Connects your MCP-compatible editor directly to Sanity's content management platform through their official API. Exposes tools for running GROQ queries, creating and patching documents, deploying schemas, managing publish states, and generating images. You can query content, push schema changes, and manipulate documents without leaving your editor. Includes comprehensive agent skills covering GROQ optimization, Visual Editing setup, Portable Text handling, and framework integrations. Works with Cursor, Claude Code, VS Code, and other MCP clients. Particularly solid if you're building content-heavy applications and want your AI assistant to understand Sanity's conventions around content modeling and performance patterns.
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →
Connect Claude to +800M contacts, +150M companies. Find & Enrich leads in chat.
Try For Free →Tools
Public tool metadata for what this MCP can expose to an agent.
41 tools
get_schemaGet the full schema of the current Sanity workspace3 params
Get the full schema of the current Sanity workspace
Parameters* required
typestringOptional: Specific type name to fetch. If not provided, returns the full schema without detailed field definitions. Full field definitions are only available when requesting a specific type.
resourceobjectResource information indicating which project id and dataset to target
workspaceNamestringWorkspace name derived from the manifest, not document type. Derived from context or listSchemaWorkspacesTooldefault: default
list_workspace_schemasGet a list of all available workspace schema names1 params
Get a list of all available workspace schema names
Parameters* required
resourceobjectResource information indicating which project id and dataset to target
deploy_schemaDeploy Sanity schema directly to cloud (project and dataset). IMPORTANT: Only use this for projects WITHOUT a local Sanity Studio. If sanity.config.ts/js or a schema directory exists, generate local schema files instead. Can be called iteratively to add/overwrite types.3 params
Deploy Sanity schema directly to cloud (project and dataset). IMPORTANT: Only use this for projects WITHOUT a local Sanity Studio. If sanity.config.ts/js or a schema directory exists, generate local schema files instead. Can be called iteratively to add/overwrite types.
Parameters* required
resourceobjectResource information indicating which project id and dataset to target
workspaceNamestringWorkspace name derived from the manifest, not document type. Derived from context or listSchemaWorkspacesTooldefault: default
schemaDeclarationstring
Sanity schema declaration
<syntax>
- Requires literal values only: no spread operators, variables, dynamic expressions, imports, or function
bodies
- If your schema makes references to other custom types in references, add those type declarations
as well, or in a separate call to this tool first
- Multiple schema types can be declared in the same source string. Simply append the objects one after another
- The schema declarations are not evaluated, so keep it declarative
</syntax>
<example>
{
name: 'author',
type: 'document',
fields: [
{
name: 'name',
type: 'string',
},
],
}
{
name: 'article',
type: 'document',
fields: [
{
name: 'title',
type: 'string',
validation: (Rule) => Rule.required().max(100),
},
{
name: 'slug',
type: 'slug'
},
{
name: 'body',
type: 'array',
of: [{type: 'block'}],
},
{
name: 'author',
type: 'reference',
to: [{type: 'author'}],
},
],
}
</example>
create_documents_from_jsonCreate one or more draft documents by directly providing JSON content. Creates drafts (drafts.* prefix) unless releaseId is specified for version creation.4 params
Create one or more draft documents by directly providing JSON content. Creates drafts (drafts.* prefix) unless releaseId is specified for version creation.
Parameters* required
resourceobjectResource information indicating which project id and dataset to target
documentsarrayArray of documents to create. Each document must have a type and content.
releaseIdstringOptional release ID for creating versioned documents. If provided, documents will be created under the specified release version instead of as drafts.
workspaceNamestringWorkspace name derived from the manifest, not document type. Derived from context or listSchemaWorkspacesTooldefault: default
create_documents_from_markdownCreate one or more draft documents from Markdown content. Creates drafts (drafts.* prefix) unless releaseId is specified for version creation.4 params
Create one or more draft documents from Markdown content. Creates drafts (drafts.* prefix) unless releaseId is specified for version creation.
Parameters* required
resourceobjectResource information indicating which project id and dataset to target
documentsarrayArray of documents to create. Each document must have a type and Markdown content.
releaseIdstringOptional release ID for creating versioned documents. If provided, documents will be created under the specified release version instead of as drafts.
workspaceNamestringWorkspace name derived from the manifest, not document type. Derived from context or listSchemaWorkspacesTooldefault: default
create_versionCreate a version document (versions.{releaseId}.* prefix) for a specific release. Versions are separate from drafts and published documents, and are used for scheduled release workflows.5 params
Create a version document (versions.{releaseId}.* prefix) for a specific release. Versions are separate from drafts and published documents, and are used for scheduled release workflows.
Parameters* required
resourceobjectResource information indicating which project id and dataset to target
releaseIdstringID of the release to associate this version with
documentIdsarrayArray of document IDs to create versions for (min 1, max 10)
instructionstringNatural language instruction for internal Sanity AI to modify the document while creating the version. Use this to make content updates, rewrites, or improvements (e.g., "Update all dates to 2025", "Make the tone more formal", "Translate technical terms to plain language"). If omitted, creates an exact copy of the source document.
workspaceNamestringWorkspace name derived from the manifest, not document type. Derived from context or listSchemaWorkspacesTooldefault: default
patch_document_from_jsonApply precise modifications to document fields. When targeting a published document, this creates or updates a draft with the changes (published document remains unchanged). When targeting a draft or version, updates it in place.7 params
Apply precise modifications to document fields. When targeting a published document, this creates or updates a draft with the changes (published document remains unchanged). When targeting a draft or version, updates it in place.
Parameters* required
setarrayBATCHED operation that replaces field values at specified paths.
Can set fields that don't exist yet - the field will be created based on schema.
CRITICAL: 'set' is DESTRUCTIVE - scope matters:
- Parent path: Replaces entire object/array item (destroys all children)
- Child path: Replaces only that field (preserves siblings)
Choose the most specific path for your intent.
Common patterns:
- To modify multiple fields: Set each field individually as SIBLINGS
Example: [{"path": "metadata.title", "value": "..."}, {"path": "metadata.description", "value": "..."}]
- To replace entire object: Set parent with ALL fields included
Example: [{"path": "metadata", "value": {"title": "...", "description": "...", "author": "..."}}]
For references: {_type: "reference", _ref: "id"}. For blocks: include _type.
unsetarrayRemoves fields and array items from the document.
Use UNSET to delete data - specify only what to remove.
Array item removal:
- Object arrays (with _key): "categories[_key==\"guide\"]", "tags[_key==\"featured\"]"
- Primitive arrays (strings, numbers): "tags[0]", "scores[2]", "categories[1]"
- Remove multiple: ["tags[_key==\"abc\"]", "items[_key==\"def\"]"]
Field removal:
- Simple fields: "metadata.draft", "tempData"
- Nested fields: "author.internalNotes"
- Remove multiple: ["metadata.title", "metadata.description"]
Can only unset fields that EXIST in the current document.
appendarrayAdds items to the END of an array.
Can ONLY append to arrays that already EXIST in the document.
CANNOT be used on non-array fields.
Each append operation targets ONE specific array.
For references: [{_type: "reference", _ref: "document-id"}]
For blocks: [{_type: "blockType", field1: "...", field2: "..."}]
resourceobjectResource information indicating which project id and dataset to target
releaseIdstringOptional release ID for patching within a specific release. Omit for standard drafts.
documentIdstringDocument ID to patch. For drafts use "drafts." prefix (e.g., "drafts.myDocId"). For published use ID directly.
workspaceNamestringWorkspace name derived from the manifest, not document type. Derived from context or listSchemaWorkspacesTooldefault: default
patch_document_from_markdownPatch a specific field in a document using Markdown content6 params
Patch a specific field in a document using Markdown content
Parameters* required
pathstringPath to the field to update (e.g., "bio", "hero.contents")
markdownstringMarkdown content to apply to the target field. The content will be converted to the appropriate structure based on the target field schema.
resourceobjectResource information indicating which project id and dataset to target
releaseIdstringOptional release ID for patching versioned documents. If provided, the document in the specified release will be patched.
documentIdstringThe ID of the document to patch
workspaceNamestringWorkspace name derived from the manifest, not document type. Derived from context or listSchemaWorkspacesTooldefault: default
query_documentsQuery documents from Sanity using GROQ query language6 params
Query documents from Sanity using GROQ query language
Parameters* required
limitintegerMaximum number of documents to returndefault: 10
querystringComplete GROQ query (e.g. "*[_type == \"post\"][0...10]{title, _id}")
paramsobjectOptional parameters for the GROQ query
singlebooleanWhether to return a single document or a listdefault: false
resourceobjectResource information indicating which project id and dataset to target
perspectivevalueOptional perspective to query from: "raw", "drafts", "published", or a release ID. Models should examine available releases and perspectives in the dataset before selecting one to ensure they are querying from the most appropriate view of the content.default: raw
generate_imageGenerate a new image in a document using AI. When targeting a published document, this creates or updates a draft with the generated image (published document remains unchanged). The draft must be published separately to make changes live. For image arrays, automatically adds...5 params
Generate a new image in a document using AI. When targeting a published document, this creates or updates a draft with the generated image (published document remains unchanged). The draft must be published separately to make changes live. For image arrays, automatically adds...
Parameters* required
resourceobjectResource information indicating which project id and dataset to target
imagePathstringPath to the image field in the document (e.g., "image", "hero.image", "images"). For arrays of images, provide the array field name and a new image will be added automatically.
documentIdstringThe ID of the document containing the image
instructionstringNatural language instructions for generating the image. Acts like a prompt telling the internal Sanity AI what to do. Describe the complete image to create (e.g., "A serene mountain landscape at dawn with misty valleys"). Keep instructions clear and specific for best results.
workspaceNamestringWorkspace name derived from the manifest, not document type. Derived from context or listSchemaWorkspacesTooldefault: default
transform_imageTransform an existing image in a document using AI. When targeting a published document, this creates or updates a draft with the transformed image (published document remains unchanged). The draft must be published separately to make changes live. Note: Image transformation i...5 params
Transform an existing image in a document using AI. When targeting a published document, this creates or updates a draft with the transformed image (published document remains unchanged). The draft must be published separately to make changes live. Note: Image transformation i...
Parameters* required
resourceobjectResource information indicating which project id and dataset to target
imagePathstringPath to the image field in the document (e.g., "image", "hero.image"). For image arrays, target with _key filtering: "images[_key==\"img1\"]".
documentIdstringThe ID of the document containing the image
instructionstringNatural language instructions for transforming the image. Acts like a prompt telling the internal Sanity AI what to do. Describe the desired changes (e.g., "Make the background more vibrant and add warm sunset lighting"). Keep instructions clear and specific for best results.
workspaceNamestringWorkspace name derived from the manifest, not document type. Derived from context or listSchemaWorkspacesTooldefault: default
get_documentFetch a single document by its exact ID. This is a direct ID lookup only - it does not search, filter, or query. Use when you have a specific document ID and need its full content.2 params
Fetch a single document by its exact ID. This is a direct ID lookup only - it does not search, filter, or query. Use when you have a specific document ID and need its full content.
Parameters* required
resourceobjectResource information indicating which project id and dataset to target
documentIdstringThe exact document ID to fetch. Must be a complete, known document ID (e.g., "movie-123" or "drafts.article-456"). This is a direct ID lookup only - it does not search, filter, or query.
publish_documentsPublish one or more draft documents to make them live2 params
Publish one or more draft documents to make them live
Parameters* required
idsarrayIDs of the documents to publish
resourceobjectResource information indicating which project id and dataset to target
unpublish_documentsUnpublish one or more published documents (moves them back to drafts)2 params
Unpublish one or more published documents (moves them back to drafts)
Parameters* required
idsarrayIDs of the documents to unpublish (published document IDs only)
resourceobjectResource information indicating which project id and dataset to target
version_replace_documentReplace the contents of a document version with contents from another document5 params
Replace the contents of a document version with contents from another document
Parameters* required
idstringID of the published document
typestringresourceobjectResource information indicating which project id and dataset to target
releaseIdstringID of the release that contains this document version
sourceDocumentIdstringID of the document to copy contents from
discard_draftsDiscard one or more draft documents (deletes drafts while keeping published documents intact). Can also discard release versions if releaseId is provided.3 params
Discard one or more draft documents (deletes drafts while keeping published documents intact). Can also discard release versions if releaseId is provided.
Parameters* required
idsarrayIDs of the documents to discard (published IDs without drafts. prefix)
resourceobjectResource information indicating which project id and dataset to target
releaseIdstringOptional release ID. If provided, discards the versions from this release instead of the draft versions.
version_unpublish_documentMark a document to be unpublished when the release is run3 params
Mark a document to be unpublished when the release is run
Parameters* required
idstringID of the published document
resourceobjectResource information indicating which project id and dataset to target
releaseIdstringID of the release that contains this document version
list_organizationsLists all organizations the user has access to in Sanity
Lists all organizations the user has access to in Sanity
No parameter schema in public metadata yet.
list_projectsLists all Sanity projects associated with your account
Lists all Sanity projects associated with your account
No parameter schema in public metadata yet.
get_project_studiosRetrieves all studio applications linked to a specific Sanity project1 params
Retrieves all studio applications linked to a specific Sanity project
Parameters* required
resourceobjectcreate_projectCreates a new Sanity project and initializes it with a dataset and API tokens4 params
Creates a new Sanity project and initializes it with a dataset and API tokens
Parameters* required
corsOriginstringCORS origin to allow, in format protocol://hostname[:port]. Use * for wildcard.
Common values:
- http://localhost:3000: For embedding Studio in Next.js, React, or similar frameworks during local development
- https://*.lovable.app: For Lovable (preview iframes and published apps)
- https://*.vusercontent.net: For v0 preview iframes
- https://*.vercel.app: For v0 published apps
Choose based on where your Studio or frontend will run. Additional CORS origins can be managed via the Sanity CLI using `sanity cors add`.
displayNamestringThe display name for the new project
organizationIdstringOrganization ID to create the project in
allowCredentialsbooleanAllow authenticated requests from this origin. Required for hosting a Studio. Defaults to truedefault: true
add_cors_originAdds CORS origin(s) to allow client-side requests to a Sanity project3 params
Adds CORS origin(s) to allow client-side requests to a Sanity project
Parameters* required
originstringCORS origin to allow, in format protocol://hostname[:port].
Wildcards (*) are supported for subdomains (e.g., https://*.example.com).
Common values:
- http://localhost:3000: Local development with Next.js, React, etc.
- http://localhost:3333: Standalone Sanity Studio
- https://*.lovable.app: Lovable preview and published apps
- https://*.vusercontent.net: v0 preview iframes
- https://*.vercel.app: v0 published apps
- https://your-domain.com: Your production domain
projectIdstringThe ID of the Sanity project to add the CORS origin to
allowCredentialsbooleanAllow authenticated requests from this origin. Required for hosting a Studio or making authenticated API requests. Defaults to true.default: true
list_datasetsLists all datasets in your Sanity project1 params
Lists all datasets in your Sanity project
Parameters* required
resourceobjectcreate_datasetCreates a new dataset with specified name and access settings3 params
Creates a new dataset with specified name and access settings
Parameters* required
aclModestringThe ACL mode for the datasetone of
private · publicresourceobjectdatasetNamestringThe name of the dataset (will be automatically formatted to match requirements)
update_datasetModifies a dataset's name or access control settings2 params
Modifies a dataset's name or access control settings
Parameters* required
aclModestringThe ACL mode for the datasetone of
private · publicresourceobjectResource information indicating which project id and dataset to target
list_releasesList content releases in Sanity, optionally filtered by state (active, scheduled, etc)2 params
List content releases in Sanity, optionally filtered by state (active, scheduled, etc)
Parameters* required
statestringFilter releases by state. Use "all" to get all non-archived releases (active, scheduled, published). Use "archived" or "deleted" to specifically list those.one of
active · scheduled · published · archived · deleted · alldefault: activeresourceobjectResource information indicating which project id and dataset to target
create_releaseCreate a new content release in Sanity with an automatically generated ID5 params
Create a new content release in Sanity with an automatically generated ID
Parameters* required
titlestringTitle for the release (e.g., "Spring 2025 Product Launch")
resourceobjectResource information indicating which project id and dataset to target
descriptionstringDescription for the release
releaseTypestringType of release (asap, undecided, scheduled)one of
asap · undecided · scheduleddefault: undecidedintendedPublishAtstringDate can be ISO format (2025-04-04T18:36:00.000Z) or natural language like "in two weeks"
edit_releaseUpdate metadata for an existing content release6 params
Update metadata for an existing content release
Parameters* required
titlestringTitle for the release (e.g., "Spring 2025 Product Launch")
resourceobjectResource information indicating which project id and dataset to target
releaseIdstringID of the release
descriptionstringDescription for the release
releaseTypestringType of release (asap, undecided, scheduled)one of
asap · undecided · scheduledintendedPublishAtstringWhen the release is intended to be published (informational only)
schedule_releaseSchedule a content release to be published at a specific time3 params
Schedule a content release to be published at a specific time
Parameters* required
resourceobjectResource information indicating which project id and dataset to target
publishAtstringDate can be ISO format (2025-04-04T18:36:00.000Z) or natural language like "in two weeks"
releaseIdstringID of the release
publish_releasePublish a release immediately2 params
Publish a release immediately
Parameters* required
resourceobjectResource information indicating which project id and dataset to target
releaseIdstringID of the release
archive_releaseArchive a release that is no longer active2 params
Archive a release that is no longer active
Parameters* required
resourceobjectResource information indicating which project id and dataset to target
releaseIdstringID of the release
unarchive_releaseRestore an archived release2 params
Restore an archived release
Parameters* required
resourceobjectResource information indicating which project id and dataset to target
releaseIdstringID of the release
unschedule_releaseRemove a previously set schedule from a release2 params
Remove a previously set schedule from a release
Parameters* required
resourceobjectResource information indicating which project id and dataset to target
releaseIdstringID of the release
delete_releaseDelete a release2 params
Delete a release
Parameters* required
resourceobjectResource information indicating which project id and dataset to target
releaseIdstringID of the release
list_embeddings_indicesList all available embeddings indices for a dataset1 params
List all available embeddings indices for a dataset
Parameters* required
resourceobjectResource information indicating which project id and dataset to target
semantic_searchPerform a semantic search on an embeddings index4 params
Perform a semantic search on an embeddings index
Parameters* required
limitintegerMaximum number of results to returndefault: 10
querystringThe search query to find semantically similar content
resourceobjectResource information indicating which project id and dataset to target
indexNamestringThe name of the embeddings index to search
migration_guideGet comprehensive guidance for migrating existing projects to Sanity. Supports general migration guides, analyzing existing content structures + designing Sanity schemas, and migrating content from various CMS platforms (like Contentful, WordPress, Strapi). Use this tool when...1 params
Get comprehensive guidance for migrating existing projects to Sanity. Supports general migration guides, analyzing existing content structures + designing Sanity schemas, and migrating content from various CMS platforms (like Contentful, WordPress, Strapi). Use this tool when...
Parameters* required
guidestringThe type of migration guide to retrieve.
- Use "general" for a high-level Sanity migration guide
- Use "schema.{platform}" for a platform-specific guide for migrating schemas to Sanity
- Use "content.{platform}" for a platform-specific guide for migrating content to Sanityone of
general · schema.generic · schema.contentful · schema.wordpress · schema.strapi · schema.drupalsearch_docsSearch Sanity docs2 params
Search Sanity docs
Parameters* required
limitintegerMaximum number of results to return (default: 10, max: 50)default: 10
querystringThe search query for documentation
read_docsRead a specific Sanity documentation article1 params
Read a specific Sanity documentation article
Parameters* required
pathstringThe documentation path (e.g., "/docs/studio/installation") or full URL (e.g., "https://www.sanity.io/docs/studio/installation")
list_sanity_rulesIMPORTANT: Call this FIRST before any Sanity development work. Returns available best practice rules for schemas, queries, frameworks, and visual editing. You MUST load relevant rules before writing Sanity code.
IMPORTANT: Call this FIRST before any Sanity development work. Returns available best practice rules for schemas, queries, frameworks, and visual editing. You MUST load relevant rules before writing Sanity code.
No parameter schema in public metadata yet.
get_sanity_rulesLoad Sanity development rules. ALWAYS call list_sanity_rules first to see available rules, then load all rules relevant to your task in a single call.1 params
Load Sanity development rules. ALWAYS call list_sanity_rules first to see available rules, then load all rules relevant to your task in a single call.
Parameters* required
rulesarrayOne or more rule names to load. Use list_sanity_rules to see available rules.
Sanity Agent Toolkit
Collection of resources to help AI agents build better with Sanity. Supports Cursor, Claude Code, Codex, VS Code, Lovable, v0, Replit, OpenCode, and any other editor/agent compatible with MCP or Agent Skills.
Features
- MCP server: Direct access to your Sanity projects (content, datasets, releases, schemas) and agent rules.
- Agent skills: Comprehensive best practices skills for Sanity development, content modeling, SEO/AEO, and experimentation. Includes 21 integration/topic guides and 26 focused best-practice rules.
- Claude Code plugin: MCP server, agent skills, and slash commands for Claude Code users. Available on the official Anthropic plugin marketplace.
- Cursor plugin: MCP server, agent skills, and commands on the Cursor Marketplace.
- Codex plugin: MCP server and agent skills for OpenAI Codex users.
Get started
Choose your path based on how you want agents to work with Sanity:
- MCP server — Give your agent always up-to-date rules and full access to your Sanity projects. No local files to maintain. Works with Cursor, VS Code, Claude Code, Lovable, v0, Replit, OpenCode, and other MCP-compatible clients.
- Agent skills — Install best practices skills for Sanity, content modeling, SEO/AEO, and experimentation. Works with Cursor, Claude Code, and any Agent Skills-compatible agent.
- Plugin — Install the Sanity plugin for Cursor or Claude Code. Bundles MCP server, agent skills, and commands.
- Manual installation — Copy the skill references locally for offline use. You'll need to update them yourself.
Option 1: Install MCP server (recommended)
Give agents direct access to Sanity projects and always up-to-date agent rules via the MCP server.
Quick install via Sanity CLI
Run in terminal to detect and configure MCP for Cursor, Claude Code and VS Code automatically:
npx sanity@latest mcp configure
Uses your logged-in CLI user for authentication — no manual tokens or OAuth needed.
Client-specific instructions
Cursor
One-click install:
Or manually: Open Command Palette (Cmd+Shift+P / Ctrl+Shift+P) → View: Open MCP Settings → + New MCP Server → add to mcp.json:
{
"mcpServers": {
"Sanity": {
"type": "http",
"url": "https://mcp.sanity.io"
}
}
}
Claude Code
Run in terminal. Authenticate with OAuth on next launch:
claude mcp add Sanity -t http https://mcp.sanity.io --scope user
Codex
Run in terminal. Authenticate with OAuth on next launch:
codex mcp add Sanity --url https://mcp.sanity.io
Or manually add to ~/.codex/config.toml:
[mcp_servers.Sanity]
url = "https://mcp.sanity.io"
VS Code
Open Command Palette (Cmd+Shift+P / Ctrl+Shift+P) → MCP: Open User Configuration → add:
{
"servers": {
"Sanity": {
"type": "http",
"url": "https://mcp.sanity.io"
}
}
}
Lovable
Sanity is available as a prebuilt chat connector in Lovable:
- Open Connectors → Chat connectors
- Select Sanity
- Click Connect and sign in to authorize your Sanity account
In your next prompt, reference your Sanity project or ask the agent to read your schema.
See the Lovable MCP documentation or Sanity + Lovable guide for more details.
v0
In the prompt input field, click Prompt Tools → MCPs → Add New → Select Sanity → Authorize → Authenticate with OAuth.
Replit
Go to Integrations Page → scroll to MCP Servers for Replit Agent → Add MCP server → Enter Sanity as name and https://mcp.sanity.io as Server URL → Test & Save → Authenticate with OAuth.
OpenCode
Add to your opencode.json:
{
"$schema": "https://opencode.ai/config.json",
"mcp": {
"sanity": {
"type": "remote",
"url": "https://mcp.sanity.io",
"oauth": {}
}
}
}
Then run: opencode mcp auth sanity
Other clients
For any MCP-compatible client, add https://mcp.sanity.io as the server URL.
If your client doesn't support remote MCP servers, use a proxy like mcp-remote:
{
"mcpServers": {
"Sanity": {
"command": "npx",
"args": ["mcp-remote", "https://mcp.sanity.io", "--transport", "http-only"]
}
}
}
Authorization and troubleshooting
Manual MCP configuration uses OAuth by default. You can use token auth instead by setting an Authorization: Bearer <token> header in the MCP config. If authentication fails after CLI setup, rerun npx sanity@latest mcp configure and restart your MCP client. For OAuth reset issues, Cursor provides Cursor: Clear All MCP Tokens and VS Code provides Authentication: Remove Dynamic Authentication Providers.
See the Sanity MCP docs for authorization options and troubleshooting.
Option 2: Install Agent Skills
Install best practices skills that work with any Agent Skills-compatible agent.
npx skills add sanity-io/agent-toolkit
See Option 3 for plugin installation.
Option 3: Install plugin
Install the Sanity plugin to get MCP server, agent skills, and commands. Available on the Claude Code marketplace and Cursor Marketplace.
Claude Code
The Sanity plugin is listed on the official Anthropic plugin marketplace. The official marketplace (claude-plugins-official) is pre-registered when you start Claude Code — you do not need to add a custom marketplace.
Install from Claude Code:
/plugin install sanity@claude-plugins-official
If the plugin is not found, refresh the marketplace catalog and retry:
/plugin marketplace update claude-plugins-official
Then run /reload-plugins to activate without restarting.
Alternative: interactive install
- Run
/pluginand open the Discover tab - Search for Sanity
- Review what the plugin will install — commands, skills, hooks, and MCP servers — before confirming (Anthropic recommends reviewing plugin permissions and source before installing)
- Choose an installation scope:
- User (default): all projects on this machine
- Project: shared with collaborators via
.claude/settings.json - Local: this repository only
- Run
/reload-pluginsto activate without restarting
Verify installation: Ask Claude Code: "which skills do you have access to?"
You should see the Sanity skills listed.
Start using: Use natural language and skills activate automatically:
Help me create a blog post schema in Sanity
Review my GROQ query and Next.js Visual Editing setup
Or run /sanity to explore all capabilities.
Cursor
Install from the Cursor Marketplace by running this in Cursor chat:
/add-plugin sanity
Verify installation: Ask Cursor: "which skills do you have access to?"
You should see the Sanity skills listed.
Start using: Use natural language and skills activate automatically:
Help me create a blog post schema in Sanity
Review my GROQ query and Next.js Visual Editing setup
Codex
- Add the Sanity marketplace:
codex plugin marketplace add sanity-io/agent-toolkit
-
Install the plugin from Codex's plugin directory (select the Sanity Agent Toolkit marketplace, then install Sanity).
-
Restart Codex. Verify by asking: "which skills do you have access to?" — you should see the Sanity skills listed.
Option 4: Manual installation
Install the skill references locally to teach your editor Sanity best practices:
- Copy
skills/sanity-best-practices/to your project. - (Recommended) Copy
AGENTS.mdto your project root to act as a knowledge router.
Capabilities
MCP tools
With MCP connected, your AI can use tools like:
query_documents— run GROQ queries directlycreate_documents— create draft documents from structured content, or version documents when a release ID is providedpatch_documents— surgical edits to existing documents; published documents are edited by creating or updating draftspublish_documents/unpublish_documents— manage document lifecycledeploy_schema/get_schema— deploy MCP-managed schemas and inspect deployed schemasdeploy_studio— deploy a hosted Studio bound to an MCP-managed schemacreate_release/list_releases— create and inspect Content Releasescreate_version— create version documents for releasesgenerate_image/transform_image— AI image generation and editingwhoami— verify the authenticated Sanity userget_project_studios— list Studio applications linked to a projectsearch_docs/read_docs— search and read Sanity documentationlist_sanity_rules/get_sanity_rules— load agent rules on demandgive_feedback— report MCP tool errors, missing capabilities, confusing output, or documentation issues
MCP-managed schemas are resolved before Studio-deployed and legacy schemas. If you deploy schema changes with deploy_schema, redeploy any matching MCP-managed Studio with deploy_studio so it picks up the latest schema. generate_image, transform_image, and create_version with an instruction consume Sanity AI credits.
See the full list of available tools.
Agent skills
Best practices skills that agents like Claude Code, Cursor, GitHub Copilot, etc. can discover and use automatically. Skills follow the Agent Skills format. See Option 2 for installation.
| Skill | Description |
|---|---|
| sanity-best-practices | GROQ performance, schema design, Visual Editing, images, Portable Text, Studio, TypeGen, localization, migrations, and framework integration guides |
| content-modeling-best-practices | Structured content principles: separation of concerns, references vs embedding, content reuse |
| seo-aeo-best-practices | SEO/AEO with EEAT principles, structured data (JSON-LD), technical SEO patterns |
| content-experimentation-best-practices | A/B testing methodology, statistical foundations, experiment design |
Getting started flow
The onboarding guide follows three phases:
- Studio & Schema — Set up Sanity Studio and define your content model
- Content — Import existing content or generate placeholder content via MCP
- Frontend — Integrate with your application (framework-specific)
Just say: "Get started with Sanity" to begin.
Slash commands (Claude Code)
| Command | What it does |
|---|---|
/sanity | List available skills and help topics |
/sanity-review | Review code for Sanity best practices |
/typegen | Run TypeGen and troubleshoot issues |
/deploy-schema | Deploy schema with verification |
Repository structure
Note: The reference files in
skills/sanity-best-practices/references/are the canonical content for the Sanity MCP server'slist_sanity_rules/get_sanity_rulestools. Each file must have validnameanddescriptionfrontmatter — rule names are derived from filenames (e.g.,nextjs.md→nextjs).
sanity-io/agent-toolkit/
├── AGENTS.md # Knowledge router & agent behavior
├── README.md # This file
├── .agents/plugins/ # Codex marketplace
│ └── marketplace.json # Codex marketplace metadata
├── .claude-plugin/ # Claude Code plugin configuration (distributed via claude-plugins-official)
│ ├── plugin.json # Plugin manifest (name: sanity)
│ └── marketplace.json # Marketplace manifest for repo-based discovery
├── .codex-plugin/ # Codex plugin configuration
│ └── plugin.json # Codex plugin manifest
├── .cursor-plugin/ # Cursor plugin configuration (distributed via cursor.com/marketplace)
│ ├── marketplace.json # Cursor marketplace metadata
│ └── plugin.json # Per-plugin manifest
├── .mcp.json # MCP server configuration
├── assets/ # Plugin branding
│ └── logo.svg # Sanity logo for marketplace display
├── commands/ # Agent commands
│ ├── sanity.md # /sanity help
│ ├── sanity-review.md # /sanity-review
│ ├── typegen.md # /typegen
│ └── deploy-schema.md # /deploy-schema
├── scripts/ # Validation and CI scripts
│ └── validate-cursor-plugin.mjs # Cursor plugin validator
└── skills/ # Agent skills (agentskills.io format)
├── sanity-best-practices/ # Comprehensive Sanity skill
│ ├── SKILL.md # Skill definition and quick reference
│ └── references/ # Canonical content (22 guides)
│ ├── get-started.md # Onboarding guide
│ ├── nextjs.md # Next.js integration
│ ├── groq.md # GROQ patterns & performance
│ ├── schema.md # Schema design & validation
│ └── ... # See SKILL.md for full index
├── content-modeling-best-practices/ # Modeling guidance + topic references
├── seo-aeo-best-practices/ # SEO/AEO guidance + topic references
└── content-experimentation-best-practices/ # Experiment design + stats references
All skills use references/ for detailed content loaded on demand. The sanity-best-practices references are also the canonical source for the MCP server's Sanity rules.
Resources
- Create Sanity account
- Sanity documentation
- GROQ language reference
- Visual Editing guide
- Sanity TypeGen
- MCP server docs
- Blueprints Infrastructure as Code
Contributing
Found a better pattern? Missing a framework or best practice? Read the contributing guide for how skills work and what makes a good contribution, then:
- Fork the repo.
- Install dependencies with
npm install. - Make your changes in
skills/<skill-name>/. - Run
npm run validate:allto check skill and plugin validity. - Submit a PR.
Support
License: MIT
Featured
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →Connect Claude to +800M contacts, +150M companies. Find & Enrich leads in chat.
Try For Free →
