generated: '2026-08-29'
method: searched
source: https://docs.soda.io/soda-ai/soda-mcp.md + https://docs.soda.io/soda-ai/soda-mcp/mcp-tools.md
provider: Soda
providerId: soda-data
name: soda-mcp
description: soda-mcp is Soda's first-party Model Context Protocol server for Soda Cloud. It runs as a
  LOCAL stdio process started by the MCP client (Claude Code, Cursor, Codex, GitHub Copilot), authenticates
  to Soda Cloud with an API key ID/secret passed through environment variables, and exposes the user-callable
  Soda Cloud Public API v4 surface as MCP tools.
deployment:
  mode: local-stdio
  install: uvx -qq --no-progress soda-mcp@latest
  package: 'soda-mcp (Soda private PyPI: team.pypi.cloud.soda.io | team.pypi.us.soda.io | enterprise.pypi.cloud.soda.io
    | enterprise.pypi.us.soda.io)'
  auth: api-key
  verified: searched
  note: 'There is NO remote/hosted MCP endpoint. Soda publishes only a stdio package, and the package
    itself is entitled: it is served from Soda''s private per-licence PyPI indexes and resolves only with
    a Soda Cloud API key in UV_INDEX, so a human with a Team or Enterprise licence must install it on
    a machine before any agent can call it. github.com/sodadata/soda-mcp returns 404 and pypi.org has
    no soda-mcp project. tools/list could not be probed for the same reason — there is no URL to POST
    to.'
transport: stdio
requirements:
- Python 3.10+
- uv on PATH
- Soda Cloud API key ID + secret
- Soda Cloud host URL (cloud.soda.io or cloud.us.soda.io)
- Team or Enterprise licence entitling the private package index
environment:
- SODA_CLOUD_HOST
- SODA_API_KEY_ID
- SODA_API_KEY_SECRET
- UV_INDEX
clients_documented:
- Claude Code
- OpenCode
- Cursor
- Codex
- GitHub Copilot
backing_api:
  name: Soda Cloud API v4
  spec: openapi/_original/soda-data-cloud-api-v4-openapi.yml
  servers:
  - https://cloud.soda.io
  - https://cloud.us.soda.io
tools_source: https://docs.soda.io/soda-ai/soda-mcp/mcp-tools.md
tools_note: 'Names and descriptions are verbatim from Soda''s published tool catalogue. inputSchema is
  NOT recorded: the live server cannot be introspected anonymously (local stdio + entitled package), so
  per-tool input schemas require an authenticated install. Each tool''s real parameters are the parameters/requestBody
  of the OpenAPI operation it is bound to in mcp/soda-data-tool-crosswalk.yml.'
summary:
  tool_count: 78
  categories: 10
  read_tools: 33
  write_tools: 45
categories:
- name: Attributes
  tool_count: 4
  tools:
  - name: list_attributes
    description: List attribute definitions available in your Soda Cloud organization for datasets, checks,
      and columns.
  - name: create_attribute
    description: Create a new attribute definition for datasets, checks, or columns.
  - name: update_attribute
    description: Update an attribute's label, description, or allowed values.
  - name: delete_attribute
    description: Delete an attribute definition. The attribute is removed from any resources where it
      is set.
- name: Checks
  tool_count: 2
  tools:
  - name: list_checks
    description: List checks in your Soda Cloud organization, including associated datasets, agreements,
      and linked incidents.
  - name: delete_check
    description: Delete a specific check by its ID. Checks are deterministic validations defined in data
      contracts (not monitors).
- name: Datasets
  tool_count: 16
  tools:
  - name: list_datasets
    description: List datasets in your Soda Cloud organization, including data source, incidents, attributes,
      and health status.
  - name: get_dataset
    description: Get a single dataset by dataset ID.
  - name: get_dataset_by_dataset_qualified_name
    description: 'Get a single dataset by dataset qualified name (DQN): [datasource _name]/ [database
      _name]/ [schema _name]/ [dataset _name].'
  - name: update_dataset
    description: Update dataset properties (label, tags, attributes, owners, profiling, metric monitoring,
      diagnostics warehouse, compute warehouse, time partition).
  - name: list_dataset_columns
    description: List active columns of a dataset with current assigned column attribute values.
  - name: set_column_attributes
    description: Set attribute values on columns of a dataset.
  - name: delete_dataset
    description: Delete a dataset. This action is permanent and cannot be undone.
  - name: list_dataset_roles
    description: List available dataset roles. Use dataset roles to manage access to individual datasets.
  - name: create_dataset_role
    description: Create a custom dataset role.
  - name: update_dataset_role
    description: Update the name or permissions of a custom dataset role.
  - name: delete_dataset_role
    description: Delete a custom dataset role. This action is permanent and cannot be undone.
  - name: get_dataset_compute_warehouse
    description: Get the compute warehouse configuration for a dataset.
  - name: get_dataset_diagnostics_warehouse
    description: Get diagnostics warehouse information for a dataset.
  - name: get_dataset_profiling
    description: Get profiling information for a dataset (structure, statistical summaries, data characteristics).
  - name: list_dataset_responsibilities
    description: List user and user group permissions assigned to a dataset, and their associated roles.
  - name: update_dataset_responsibilities
    description: Update user/group permissions and their associated roles for a dataset.
- name: Monitors
  tool_count: 8
  tools:
  - name: get_dataset_metric_monitoring
    description: Get metric monitoring configuration for a dataset, including all three monitor types
      and the dataset-level scan schedule.
  - name: create_column_metric_monitor
    description: Create a user-defined column metric monitor for a dataset.
  - name: update_column_metric_monitor
    description: Update an existing user-defined column metric monitor (e.g. sensitivity, thresholds,
      grouping, exclusion zones).
  - name: delete_column_metric_monitor
    description: Delete a user-defined column metric monitor. This action is permanent and cannot be undone.
  - name: create_custom_sql_monitor
    description: Create a user-defined custom SQL monitor for a dataset.
  - name: update_custom_sql_monitor
    description: Update an existing user-defined custom SQL monitor (e.g. SQL query, sensitivity, thresholds,
      grouping, exclusion zones).
  - name: delete_custom_sql_monitor
    description: Delete a user-defined custom SQL monitor. This action is permanent and cannot be undone.
  - name: run_historical_metric_collection
    description: Trigger a historical metric collection scan for a dataset. The v4 API calls this a historical
      metric collection scan; in user-facing terms it backfills metric history so monitors (ML-based anomaly
      detection) have a baseline to learn from.
- name: Contracts
  tool_count: 10
  tools:
  - name: list_contracts
    description: List data contracts in your Soda Cloud organization. A data contract is the declarative
      source-of-truth for checks on a dataset.
  - name: create_contract
    description: Create a new data contract on a dataset. The contract is initialized with the YAML contents
      you supply; the full contract YAML is normally installed later via `publish_contract`.
  - name: get_contract
    description: Retrieve a specific data contract by ID, including its YAML content.
  - name: publish_contract
    description: Publish new YAML content for an existing data contract.
  - name: list_contract_versions
    description: List published versions of a specific data contract. Each publish creates a new version;
      this tool returns the version history.
  - name: verify_contract
    description: Trigger a contract verification scan for the given contract. Runs the checks defined
      in the contract against the dataset.
  - name: create_skeleton_contract
    description: Trigger async generation of a skeleton contract for a dataset, derived from its warehouse
      schema. Use this to bootstrap a contract's YAML body when no contract has been published yet.
  - name: get_skeleton_contract_status
    description: Get the status of a skeleton contract generation operation started via `create_skeleton_contract`.
  - name: generate_contracts
    description: Trigger async AI-powered generation of full contracts for one or more datasets.
  - name: get_contract_generation_status
    description: Get the status of a contract generation operation started via `generate_contracts`.
- name: Data Standards
  tool_count: 11
  tools:
  - name: list_data_standards
    description: List data standards in your Soda Cloud organization. A data standard is an organization-wide
      YAML policy whose checks are automatically applied to every dataset matching its scope.
  - name: get_data_standards_activity
    description: Get an organization-wide rollup of data standards activity, including counts of active
      and total standards, matched datasets, and aggregated check results.
  - name: get_data_standard
    description: Retrieve a specific data standard by ID, including its YAML contents and scope.
  - name: list_data_standard_checks
    description: List the aggregated checks generated by a specific data standard across the datasets
      it matches.
  - name: list_data_standard_datasets
    description: List the datasets currently matched by a specific data standard's scope.
  - name: evaluate_data_standard_scope
    description: Re-evaluate which datasets currently match a data standard's scope and return their dataset
      IDs.
  - name: create_data_standard
    description: Create a new data standard from YAML contents, a structured scope, status, owners, and
      an optional schedule. A data standard is an organization-wide YAML policy whose checks are automatically
      applied to every dataset matching its scope.
  - name: update_data_standard
    description: Update an existing data standard's YAML contents, scope, status, owners, and schedule.
  - name: update_data_standard_status
    description: Transition a data standard to a new status without modifying its contents, scope, owners,
      or schedule. Use this to pause or activate an existing standard.
  - name: delete_data_standard
    description: Delete a data standard.
  - name: execute_data_standards
    description: Trigger a scan that runs the active data standards linked to a dataset. Identify the
      dataset by `datasetId`.
- name: Datasources
  tool_count: 18
  tools:
  - name: list_datasources
    description: List datasources in your Soda Cloud organization. A datasource is a configured connection
      to a warehouse (e.g. Snowflake, Postgres) that owns datasets.
  - name: run_datasource_discovery
    description: Trigger a discovery scan on a datasource. Detects new tables and schemas in the warehouse
      on-demand (instead of waiting for the scheduled cron).
  - name: create_datasource
    description: Create a new datasource from a YAML configuration document. The datasource type (e.g.
      Snowflake, Postgres) is extracted from the YAML contents.
  - name: get_datasource
    description: Get a datasource's full configuration by ID, including its YAML configuration, label,
      runner, and type.
  - name: update_datasource
    description: Update a datasource's configuration. Only the fields you supply are updated (PATCH semantics);
      omitted fields keep their existing values.
  - name: delete_datasource
    description: Delete a datasource and all its associated resources (datasets, checks, scans, incidents).
  - name: list_datasource_roles
    description: List available datasource roles. Use datasource roles to manage access to individual
      datasources.
  - name: create_datasource_role
    description: Create a custom datasource role.
  - name: update_datasource_role
    description: Update the name or permissions of a custom datasource role.
  - name: delete_datasource_role
    description: Delete a custom datasource role. This action is permanent and cannot be undone.
  - name: get_datasource_diagnostics_warehouse
    description: Get the diagnostics warehouse configuration for a datasource. The diagnostics warehouse
      collects scan-related data (failed rows, scan results) and forwards it to the customer's warehouse
      for storage and analysis.
  - name: update_datasource_diagnostics_warehouse
    description: Update the diagnostics warehouse configuration for a datasource. The diagnostics warehouse
      collects scan-related data (failed rows, scan results) and stores it in your warehouse for analysis.
  - name: list_datasource_responsibilities
    description: List user and user group permissions assigned to a datasource, and their associated roles.
  - name: update_datasource_responsibilities
    description: Update user/group permissions and their associated roles for a datasource.
  - name: test_datasource_connection
    description: Trigger an async connection test for a datasource configuration. Use this to validate
      a YAML + runner combination without creating a datasource, e.g. before calling `create_datasource`.
  - name: get_datasource_connection_test_status
    description: Get the status of an async datasource-connection-test operation started via `test_datasource_connection`.
  - name: onboard_discovered_datasets
    description: Trigger async onboarding of one or more discovered datasets for a datasource. Onboarding
      promotes a discovered-but-not-yet-tracked dataset into a regular Soda dataset that can carry contracts,
      checks, and monitors.
  - name: get_onboard_discovered_datasets_status
    description: Get the status of an async dataset-onboarding operation started via `onboard_discovered_datasets`.
- name: Discovered Datasets
  tool_count: 1
  tools:
  - name: list_discovered_datasets
    description: List datasets Soda has discovered in your data sources. Discovered datasets have been
      found during a discovery scan but may not yet be onboarded as first-class Soda datasets.
- name: Runners
  tool_count: 4
  tools:
  - name: list_runners
    description: List Soda runners in your organization.
  - name: create_runner
    description: Create API key credentials for a new self-hosted Soda runner deployment.
  - name: get_runner
    description: Get a runner by ID, including online status, runner type, version information, and last-seen
      timestamp.
  - name: delete_runner
    description: Delete a self-hosted runner by ID only after the user has explicitly confirmed deletion.
- name: Secrets
  tool_count: 4
  tools:
  - name: list_secrets
    description: List secrets in your organization.
  - name: create_secret
    description: Create a new secret that can be referenced in datasource YAML as `${secret.NAME}`.
  - name: delete_secret
    description: Delete an existing secret.
  - name: update_secret
    description: Update the value of an existing secret. The secret name cannot be changed.
related:
  claude_code_plugin: https://github.com/sodadata/soda-agentic-tools
  product_page: https://www.soda.io/product/mcp