Can AI agents automate Databricks Data Intelligence Platform?
Databricks · saas · databasebusiness-intelligencesaas-platform
Databricks is automatable today through workspace REST APIs and the SQL Statement Execution API at POST /api/2.0/sql/statements, which Databricks marks generally available. Agents should prefer OAuth: user-to-machine for interactive work and service-principal client-credentials for unattended jobs. Personal access tokens remain documented as a legacy workspace-only fallback, capped at 600 per user and revoked after 90 days unused. Results can be inline JSON (25 MiB) or external links in JSON, CSV, or Arrow. Official SDKs exist for Python, JavaScript, Go, and Java. The official Databricks CLI (v1 and later) wraps the same REST surface. First-party managed MCP servers are hosted in the workspace under Unity AI Gateway and Unity Catalog permissions. They expose Genie, AI Search, Databricks SQL, and Unity Catalog functions over /api/2.0/mcp/... with OAuth scopes. The feature is Public Preview as of the 2026-08-17 docs update. There is no shared public MCP URL. Snowflake, dbt, and Looker are outside this record. Computer-use viability is unassessed. API and CLI both score 8; API is the best path because it is the machine contract and MCP is still preview.
Best path today: api · Overall automatability: 8/10
Modalities
| Modality | Status | Detail |
|---|---|---|
| API | yes | restoauth2-u2moauth2-m2mpersonal-access-token coverage: partial · docs |
| SDK | yes | pythonjavascriptgojava official: yes |
| MCP | yes | verdict: official |
| Integrations | unknown | |
| CLI | yes | |
| Extensibility | yes | sqlnotebooksunity-catalog-functions webhooks: unknown |
| Data access | export: json, csv, arrow · import: sql | |
| RPA / UI automation | unknown | web-dom No repeatable UI probe was run. Workspace REST, SQL Statement Execution, the official CLI, and managed MCP servers cover warehouse and catalog work without browser control.
|
| Computer use | unknown | measured verdicts only — "unknown" means not yet probed by us |
Scores
| api | mcp | integrations | cli | rpa | computer use | overall |
|---|---|---|---|---|---|---|
| 8 | 6 | — | 8 | — | — | 8 |
Evidence
| Fact | Tier | Source | Date |
|---|---|---|---|
homepage | declared | source | 2026-08-30 |
license | declared | source | 2026-08-30 |
modalities.api.exists | declared | source | 2026-08-30 |
modalities.api.kinds[0] | declared | source | 2026-08-30 |
modalities.api.auth[0] | declared | source | 2026-08-30 |
modalities.api.auth[1] | declared | source | 2026-08-30 |
modalities.api.auth[2] | declared | source | 2026-08-30 |
modalities.api.coverage | declared | source | 2026-08-30 |
modalities.sdk.exists | declared | source | 2026-08-30 |
modalities.mcp.first_party | declared | source | 2026-08-30 |
modalities.cli.exists | declared | source | 2026-08-30 |
modalities.extensibility.scripting[0] | declared | source | 2026-08-30 |
modalities.extensibility.scripting[2] | declared | source | 2026-08-30 |
modalities.data_access.export[0] | declared | source | 2026-08-30 |
modalities.data_access.import[0] | declared | source | 2026-08-30 |
modalities.agent_docs.llms_txt | declared | source | 2026-08-30 |
verdict.scores.api | declared | source | 2026-08-30 |
verdict.scores.mcp | declared | source | 2026-08-30 |
verdict.scores.cli | declared | source | 2026-08-30 |
freshness.watch[0].url | declared | source | 2026-08-30 |
freshness.watch[1].url | declared | source | 2026-08-30 |
freshness.watch[2].url | declared | source | 2026-08-30 |
freshness.watch[3].url | declared | source | 2026-08-30 |
Related tools
Other products in this database that share a category with Databricks Data Intelligence Platform.
Last verified 2026-08-30 · volatility high · JSON record