Menu

Mattering Automations MCP tools

22 tools on the Mattering MCP gateway at https://mcp.mattering.io/mcp. Each call reaches the Automations service tier with your identity in headers, so a tool sees exactly what you can see in the app.

Connect any MCP client to https://mcp.mattering.io/mcp; it discovers the Mattering sign-in and asks you to approve the tools it may use. Tool names, parameters and descriptions below are read from the gateway source. See the MCP gateway concept page for connection details.

Automation workspaces

automations_create_workspace

Create a new workspace in the user's organization.

Full description

Create a new workspace in the user's organization.

Args:
    name: Human-readable name (unique among active workspaces).
    description: Optional description.
    space_id: Optional Productivity Space public_id to link.

Parameters

NameTypeRequiredDefault
namestrrequirednone
descriptionOptional[str]optional''
space_idOptional[str]optional''

Calls

  • POST/

automations_get_workspace

Retrieve a workspace by ID.

Full description

Retrieve a workspace by ID.

Returns workspace metadata, org, optional space_id link, and status.

Parameters

NameTypeRequiredDefault
workspace_idstrrequirednone

Calls

  • GET/{workspace_id}/

automations_list_workspaces

List workspaces in the user's organization.

Full description

List workspaces in the user's organization.

Optional filter: status ('active' | 'archived'; default 'active'). Returns {workspaces: [...]}.

Parameters

NameTypeRequiredDefault
statusOptional[str]optional'active'

Calls

  • GET/

Canvas Automations

automations_create_automation

Create or upsert a canvas automation workflow from a node/edge graph.

Full description

Create or upsert a canvas automation workflow from a node/edge graph.

This is how you build a workflow that appears in the Automations canvas without importing static JSON. Idempotent: with automation_id it updates that automation; otherwise it replaces an existing automation matching (org, space_id, name), else creates a new one.

Args:
    space_id: Productivity space public_id the automation belongs to.
    name: Workflow name (also the upsert key when automation_id is omitted).
    nodes: List of nodes. Each: ``{"id": "trigger", "type":
        "trigger_task_updated", "label": "...", "position": {"x": 320,
        "y": 40}, "config": {...}}. type`` is the node_type from
        automations_list_node_types (e.g. 'video.assemble_timeline',
        'function_update_custom_field'). For secrets in config use a
        {"$secret": "<uuid>"} marker: never a raw value.
    edges: List of connections. Each: ``{"source": "trigger", "target":
        "extract", "sourceHandle": "output", "targetHandle": "input"}``.
    description: Optional workflow description.
    status: 'draft' (default) | 'active' | 'paused' | 'archived'. Only
        'active' automations fire on triggers: keep 'draft' until the
        config (secrets, inputs) is complete.
    automation_id: Optional existing automation id to update in place.

Returns {success, id, name, status, created, node_count}.

Parameters

NameTypeRequiredDefault
space_idstrrequirednone
namestrrequirednone
nodeslistrequirednone
edgesOptional[list]optionalnone
descriptionOptional[str]optional''
statusOptional[str]optional'draft'
automation_idOptional[str]optionalnone

Calls

  • POST/

automations_delete_automation

Delete a canvas automation workflow by id.

Parameters

NameTypeRequiredDefault
automation_idstrrequirednone

Calls

  • DELETE/{automation_id}/

automations_get_automation

Fetch one automation with its full node/edge graph (canvas format).

Full description

Fetch one automation with its full node/edge graph (canvas format).

Returns ``{id, name, description, status, is_active, space_id, nodes: [{id, type, label, position, config}], edges: [{source, target, sourceHandle, targetHandle, label}]}``.

Parameters

NameTypeRequiredDefault
automation_idstrrequirednone

Calls

  • GET/{automation_id}/

automations_list_automations

List canvas automation workflows in the user's organization.

Full description

List canvas automation workflows in the user's organization.

Args:
    space_id: Optional Productivity space public_id to scope to.
        Omit to list all automations in the org.

Returns ``{automations: [{id, name, status, is_active, space_id, node_count, updated_at}], count}``.

Parameters

NameTypeRequiredDefault
space_idOptional[str]optionalnone

Calls

  • GET/

Data Studio: Entries

automations_create_entries_bulk

Bulk-insert up to 10,000 entries into a datatable in a single call.

Full description

Bulk-insert up to 10,000 entries into a datatable in a single call.

Prefer this over looping automations_create_entry when importing large datasets. Invalid rows are skipped and reported in the response; valid rows are inserted in server-side batches.

Args:
    datatable_id: UUID of the target datatable (must be 'active').
    entries: list of entry objects. Each must be a dict with a data
        key containing field values, e.g.:
        [{"data": {"name": "Alice", "age": 30}}, {"data": {"name": "Bob", "age": 25}}]
Field values follow the same validation rules as
        automations_create_entry.
Returns:
    JSON with:
    - created (int): number of entries successfully inserted.
    - error_count (int): number of entries that failed validation.
    - errors (list): first 100 validation errors, each with
      index (0-based row position) and error (description).

Parameters

NameTypeRequiredDefault
datatable_idstrrequirednone
entrieslist[dict]requirednone

automations_create_entry

Insert a new entry into a datatable.

Full description

Insert a new entry into a datatable.

Args:
    datatable_id: UUID of the target datatable (must be 'active').
    data: dict of field values. Keys must match the datatable's field
        definitions. Required fields must be present unless they have
        a default_value. Values are type-validated against field_type.
Validation rules by field type:
    string  : must be str
    number  : must be int or float
    integer : must be int
    boolean : must be bool
    datetime: must be ISO 8601 string
    json    : any JSON-serializable value
    relation: must be a valid UUID string

Parameters

NameTypeRequiredDefault
datatable_idstrrequirednone
datadictrequirednone

Calls

  • POST/datatables/{datatable_id}/entries/

automations_delete_entry

Permanently delete an entry. This is irreversible.

Full description

Permanently delete an entry. This is irreversible.

Only works on entries in 'active' datatables.

Parameters

NameTypeRequiredDefault
entry_idstrrequirednone

Calls

  • DELETE/entries/{entry_id}/

automations_get_entry

Retrieve a single entry by its ID.

Full description

Retrieve a single entry by its ID.

Returns the entry's data, datatable_id, creation/update timestamps, and lineage fields (created_by_execution, created_by_agent, etc.).

Parameters

NameTypeRequiredDefault
entry_idstrrequirednone

Calls

  • GET/entries/{entry_id}/

automations_list_entries

Query entries in a datatable with optional filters, sorting, and pagination.

Full description

Query entries in a datatable with optional filters, sorting, and pagination.

Args:
    datatable_id: UUID of the datatable.
    filters: JSON string of filter list. Each filter is a dict with:
        - field: name of the data field (or 'created_at'/'updated_at').
        - operator: 'eq', 'neq', 'gt', 'gte', 'lt', 'lte',
          'contains', 'in', 'isnull'.
        - value: comparison value.
      Example: '[{"field": "status", "operator": "eq", "value": "active"}]'
    sort_by: field name to sort by (default: 'created_at').
    sort_order: 'asc' or 'desc' (default: 'desc').
    page: page number (default: 1).
    page_size: entries per page (default: 50, max: 500).

Parameters

NameTypeRequiredDefault
datatable_idstrrequirednone
filtersOptional[str]optionalnone
sort_byOptional[str]optionalnone
sort_orderOptional[str]optional'desc'
pageOptional[int]optional1
page_sizeOptional[int]optional50

Calls

  • GET/datatables/{datatable_id}/entries/

automations_update_entry

Update an entry's data (merge-patch semantics).

Full description

Update an entry's data (merge-patch semantics).

The provided data dict is merged into the existing entry data: new keys are added, existing keys are overwritten, omitted keys are preserved. The merged result is then validated against the datatable's field definitions.

Only works on entries in 'active' datatables.

Parameters

NameTypeRequiredDefault
entry_idstrrequirednone
datadictrequirednone

Calls

  • PATCH/entries/{entry_id}/

Data Studio: Tables

automations_archive_datatable

Soft-delete (archive) a datatable. Archived tables are read-only.

Full description

Soft-delete (archive) a datatable. Archived tables are read-only.

Data is retained and can be read, but new entries cannot be created. To restore, use automations_update_datatable with status='active'.

Parameters

NameTypeRequiredDefault
datatable_idstrrequirednone

Calls

  • DELETE/datatables/{datatable_id}/

automations_create_datatable

Create a new datatable with a typed schema.

Full description

Create a new datatable with a typed schema.

Args:
    name: Human-readable name (unique within the organization).
    fields: List of field definitions. Each field dict must include:
        - name (str, required): field name used as the key in entry data.
        - field_type (str): one of 'string', 'number', 'integer',
          'boolean', 'datetime', 'json', 'relation'. Defaults to 'string'.
        - required (bool): whether entries must include this field.
          Defaults to false.
        - default_value (any): applied when the field is omitted on write.
        - description (str): human-readable description.
        - related_datatable_id (str): required for 'relation' fields : 
          UUID of the datatable whose entries this field references.
    workspace_id: Alphanumeric ID of the workspace to create the datatable in.
    description: Optional description of the datatable.

Parameters

NameTypeRequiredDefault
namestrrequirednone
fieldslist[dict]requirednone
workspace_idOptional[str]optionalnone
descriptionOptional[str]optional''

Calls

  • POST/datatables/
  • POST/{workspace_id}/datatables/

automations_get_datatable

Retrieve a datatable by ID, including its field definitions.

Full description

Retrieve a datatable by ID, including its field definitions.

Returns the datatable metadata, status, and full fields array.

Parameters

NameTypeRequiredDefault
datatable_idstrrequirednone

Calls

  • GET/datatables/{datatable_id}/

automations_list_datatables

List datatables: optionally scoped to a workspace.

Full description

List datatables: optionally scoped to a workspace.

Args:
    workspace_id: Alphanumeric ID of a workspace to filter by. When provided,
        uses the workspace-nested endpoint. Otherwise lists all
        org datatables.
    status: 'active' | 'archived' (default 'active').
Returns {results: [...], count: N}.

Parameters

NameTypeRequiredDefault
workspace_idOptional[str]optionalnone
statusOptional[str]optional'active'

Calls

  • GET/datatables/
  • GET/{workspace_id}/datatables/

automations_update_datatable

Update a datatable's metadata or schema.

Full description

Update a datatable's metadata or schema.

Updatable fields in payload:
    - name (str): new name.
    - description (str): new description.
    - status (str): 'active' or 'archived'.
    - workspace_id (str): link the table to an automations
      workspace so it appears in that workspace's Data Studio; pass
      ''/null to unlink (org-scoped). Must be in the same org.
    - fields (list): full replacement of field definitions. When
      provided, existing fields are deleted and replaced.

Parameters

NameTypeRequiredDefault
datatable_idstrrequirednone
payloaddictrequirednone

Calls

  • PATCH/datatables/{datatable_id}/

Diagnostics

automations_list_capabilities

List the automations service's registered capabilities (debug helper).

Full description

List the automations service's registered capabilities (debug helper).

Returns the available capability names, HTTP methods, paths, categories, and required / optional parameter lists. Useful for discovering what the service exposes without paging through the docs.

Takes no parameters.

Node Execution

automations_execute_node

Execute a single automation node directly (without a workflow graph).

Full description

Execute a single automation node directly (without a workflow graph).

This invokes any registered node handler: the same code that runs inside an automation workflow, but as a standalone one-shot call. Use automations_list_node_types to discover available node types and their expected config keys.

Args:
    node_type: The node type identifier, e.g. 'function_create_task',
        'action_slack_send_message', 'function_http_request',
        'action_openai_chat_completion', 'query_task', etc.
    config: Node-specific configuration dict. Keys vary by node type.
        For API keys and passwords, **never put raw secret values here**.
        Call automations_list_secrets first, then reference a secret
        by id::

{"anthropic_api_key": {"$secret": "<uuid>", "_label": "My key"}}

The server resolves markers at execution time using the
        caller's access grants; values are redacted from responses.

Common examples:

**function_create_task**::

{"title": "...", "description": "...", "widget_id": "...",
             "status": "...", "assignee_id": "..."}

**function_http_request**::

{"url": "https://...", "method": "GET",
             "headers": [{"name": "Authorization", "value": {"$secret": "<uuid>", "_label": "token"}}]}

**seo.fetch_keyword_universe**::

{"dataforseo_login": {"$secret": "<uuid>", "_label": "DF login"},
             "anthropic_api_key": {"$secret": "<uuid>", "_label": "Claude"}}

**action_openai_chat_completion**::

{"model": "gpt-4o", "messages": [{"role": "user", "content": "..."}]}

**action_slack_send_message**::

{"channel": "#general", "text": "Hello from automations"}

**query_task**::

{"task_id": "abc123"}

Values support {{nodes["nodeId"].field}} template syntax
        for referencing prior node outputs (in workflow context).
space_id: Productivity workspace/space ID for context. Required
        for productivity nodes (task/batch operations). Optional
        for global nodes (HTTP, AI, etc.).

Returns the node handler's output as JSON. Error responses include an error key with a description.

Long-running worker nodes (the seo.*, research.*, and editorial video.* families can take several minutes of external API + LLM passes) support async_mode: true in config: the call returns a job_id immediately and the pack's <prefix>.get_job_result node (seo.get_job_result, research.get_job_result, video.get_job_result) fetches the outcome. Prefer that over holding this call open when a run may exceed a few minutes.

Parameters

NameTypeRequiredDefault
node_typestrrequirednone
configdictrequirednone
space_idOptional[str]optional''

Calls

  • POST/node-exec/

automations_list_node_types

List all available automation node types grouped by category.

Full description

List all available automation node types grouped by category.

Returns the full catalog of triggers, actions, queries, and control-flow nodes that can be executed via automations_execute_node.

Args:
    scopes: Comma-separated scope filter. Common scopes:
        'global' (HTTP, AI, control-flow nodes),
        'productivity' (task/batch/notification nodes).
        Connector-gated nodes (Slack, GitHub, Stripe, etc.) are
        auto-included when the org has the connector configured.
        Default: 'global,productivity'.

Returns JSON {node_types: {category: [{node_type, label, description, ...}]}}.

Parameters

NameTypeRequiredDefault
scopesOptional[str]optional'global,productivity'

Calls

  • GET/node-palette/

Secrets

automations_list_secrets

List automation secrets the current user may reference in node config.

Full description

List automation secrets the current user may reference in node config.

Returns metadata only (id, name, integration_type, fingerprint, status). Secret **values are never included**: use the returned id in automations_execute_node config as a $secret marker (see below).

Visibility matches the automations UI picker:
  - Same organization as the authenticated user.
  - Private secrets: only if you created them.
  - Org-shareable secrets: if you created them, have a user grant, or
    the secret has an org-wide grant.
Args:
    integration_type: Optional filter (slack, anthropic, custom, etc.).
    credential_type: Optional typed-credential slug filter.
    active_only: When true (default), only active secrets: suitable
        for wiring into node configs. When false, also returns
        pending_rotation and suspended rows for admin-style tooling.

Returns {actor: {user_id, org_id}, results: [...], count: N}. Compare actor.org_id to the org shown in the automations UI (session / org switcher). If they differ, re-authorize the MCP in Cursor and pick the correct org on the consent screen.

Each result includes id for use in config markers.

Parameters

NameTypeRequiredDefault
integration_typeOptional[str]optionalnone
credential_typeOptional[str]optionalnone
active_onlybooloptionalTrue

Calls

  • GET/

Identity tools (every app)

The gateway registers these once, outside any app family. Call whoami first when a result looks empty.

auth_statusReport token health, expiry, bound principal, and catalog version.
get_accountReturn the current user's account profile (email, name, id).
get_organizationReturn the organization this session is currently acting as.
list_organizationsList every organization the authenticated identity belongs to.
reauthenticateExplain how to refresh this connection's authentication.
switch_accountExplain how to re-authenticate as a different account / org.
whoamiShow who this MCP session is authenticated as.
Start Free Explore our Products