Directus MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| DIRECTUS_URL | No | The URL of your Directus instance (e.g., https://your-directus-instance.com) | |
| MCP_TOOLSETS | No | A comma-separated list of toolsets to expose (e.g., collections,fields,relations,schema,content,flow,dashboards,all). Default is 'default' which includes collections, fields, relations, and content. | default |
| DIRECTUS_EMAIL | No | Email address for Directus authentication (Alternative to static token) | |
| DIRECTUS_TOKEN | No | Static token for Directus authentication (Recommended for production) | |
| DIRECTUS_PASSWORD | No | Password for Directus authentication (Alternative to static token) |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| list_collectionsB | List all collections in the Directus instance. Returns collection names, metadata, and schema information. |
| get_collectionB | Get detailed information about a specific collection including all fields, metadata, and schema configuration. |
| create_collectionA | Create a new collection (database table) in Directus. Automatically creates a proper database table with schema. Can include initial fields. Example: {collection: "articles", meta: {icon: "article", note: "Blog articles"}, fields: [{field: "id", type: "integer", schema: {is_primary_key: true, has_auto_increment: true}}, {field: "title", type: "string"}]} |
| update_collectionC | Update collection metadata such as icon, note, visibility settings, etc. |
| delete_collectionA | Delete a collection and all its data. This action cannot be undone. Use with caution. |
| list_fieldsB | List all fields in a specific collection with their types, metadata, and schema configuration. |
| create_fieldA | Add a new field to a collection. Specify field type, interface, and constraints. Example: {collection: "articles", field: "status", type: "string", meta: {interface: "select-dropdown", options: {choices: [{text: "Draft", value: "draft"}, {text: "Published", value: "published"}]}, required: true}} |
| update_fieldC | Update field properties such as metadata, interface options, or schema constraints. |
| delete_fieldA | Remove a field from a collection. This will delete the column and all its data. Use with caution. |
| list_relationsA | List all relations (foreign keys, M2O, O2M, M2M) in the Directus instance. |
| create_relationA | Create a relation between collections (M2O, O2M, or M2M). For M2O: specify collection, field, and related_collection. For O2M: also include meta.one_field. Example M2O: {collection: "articles", field: "author", related_collection: "users"} |
| delete_relationB | Delete a relation. Specify the collection and field that contains the relation. |
| query_itemsA | Query items from a collection with advanced filtering, sorting, pagination, and search. Supports Directus filter operators like _eq, _neq, _lt, _lte, _gt, _gte, _in, _nin, _null, _nnull, _contains, _ncontains, _starts_with, _nstarts_with, _ends_with, _nends_with, _between, _nbetween. Example: {collection: "articles", filter: {"status": {"_eq": "published"}, "date_created": {"_gte": "2024-01-01"}}, sort: ["-date_created"], limit: 10} |
| get_itemA | Get a single item by ID from a collection. Optionally specify fields to return and deep query for relational data. |
| create_itemA | Create a new item in a collection. Provide the item data as key-value pairs. Example: {collection: "articles", data: {title: "My Article", status: "draft", body: "Article content..."}} |
| update_itemA | Update an existing item in a collection. Provide the item ID and fields to update. Example: {collection: "articles", id: 1, data: {status: "published"}} |
| delete_itemA | Delete an item from a collection by ID. This action cannot be undone. |
| bulk_create_itemsA | Create multiple items in a collection at once. More efficient than creating items one by one. Example: {collection: "articles", items: [{title: "Article 1", status: "draft"}, {title: "Article 2", status: "draft"}]} |
| bulk_update_itemsA | Update multiple items in a collection at once. Each item must include an id field. Example: {collection: "articles", items: [{id: 1, status: "published"}, {id: 2, status: "published"}]} |
| bulk_delete_itemsA | Delete multiple items from a collection at once by their IDs. This action cannot be undone. Example: {collection: "articles", ids: [1, 2, 3]} |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 20 tools
Each tool has a distinct purpose: create vs update vs delete for different resources (collections, fields, items, relations), plus bulk variants and querying. No overlap that could confuse an agent.
All tools follow a consistent verb_noun pattern with snake_case (e.g., create_collection, list_fields, bulk_create_items). Prefixes like bulk_ are applied uniformly. No mixing of conventions.
20 tools cover the core operations of a headless CMS: CRUD for collections, fields, items, relations, plus bulk and query. Slightly on the high side but still well-scoped for the domain.
Covers CRUD for all main entities plus bulk operations and advanced querying. Minor gaps: no explicit list_items (query_items serves that role), no update for relations, and missing some higher-level features like user management, but core data operations are complete.