← Back to Claude Tool Reviews

Cowork Data Plugin

Connector Data & Analytics Solid

SQL query writing, statistical analysis, and dashboard creation. Connectors: Snowflake, Databricks, BigQuery, Definite, Hex, Amplitude. Rated highest-tier by one reviewer who reported it found a $14K/month pricing anomaly missed by a client's own data team.

10,917 starsApache-2.0 (commercial OK)PaidNo code needed
Built and maintained by Anthropic, the makers of Claude.

One reviewer rated it highest-tier (highest) and reported it identified a $14K/month pricing anomaly a client's own data team had missed. Strongest for ad-hoc SQL analysis and statistical work.

Data analysts and teams with Snowflake, Databricks, or BigQuery. Works with exported CSVs for teams without data warehouse connectors.

Claude Code Claude Cowork Claude Chat

https://github.com/anthropics/knowledge-work-plugins/tree/ma...

By Anthropic

How to Get It

Option 1: Claude Desktop AppOpen the Customize panel in the sidebar → browse connectors → search and add. Works in Claude Code, Claude Cowork, and Claude Chat.
Option 2: Paste into Claude CodeCopy the command below and paste it into a Claude Code conversation. Claude will run it for you.
Command
claude plugin install data@knowledge-work-plugins  OR  install via claude.com/plugins

Tip: Paste this into a Claude Code conversation. Verify command matches your Claude Code version.

PrerequisitesClaude Cowork (Pro/Max/Team/Enterprise). Snowflake, BigQuery, Databricks, or CSV files.Time to functional~5 min for CSV; ~30 min for warehouse connectionCostPaid — Requires Claude Pro ($20/mo) or higher

Trust Signals Source Reviewed

Stars10,917Contributors16Last updated2026-04-04LicenseApache-2.0 (OK for commercial use)Known CVEsNone found

Data & Access

Data processingProcessed by Anthropic (the makers of Claude)

Reviewer notes

Source Reviewed review. These are observations, not a security certification.

Official Anthropic plugin. Connectors: Snowflake, Databricks, BigQuery, Definite, Hex, Amplitude, Jira. Works with plain CSV too — low barrier entry point. /data:explore is excellent for one-off analysis. Strong client-readiness because it democratises SQL.

Anthropic-official knowledge-work-plugins repo. 10.8K stars. Connectors are account-scoped, not workspace-scoped (JJ Englert finding). This means connectors added in one workspace are accessible across all workspaces under the same Anthropic account.

2026-04-02: LIMITATION: No connector for dbt, Redshift, or Looker out of the box — the three most common alternatives at enterprises. Data teams on those stacks need custom MCP connectors.

2026-03-29: Connectors: Snowflake, Databricks, BigQuery, Definite, Hex, Amplitude, Jira. Official repo: https://github.com/anthropics/knowledge-work-plugins/tree/main/data

2026-03-29: SECURITY FINDING (JJ Englert, Mar 2026): Cowork connectors are scoped at the Anthropic account level, NOT the workspace level. A connector added in one Cowork workspace is accessible in ALL workspaces under the same Anthropic account. This means: (1) connectors with sensitive access (email, calendar, CRM) bleed across all projects/workspaces; (2) multi-tenant setups sharing an Anthropic account are NOT isolated at the connector level; (3) evaluate connector blast radius across the full account, not just the workspace being configured. Source: https://github.com/JJenglert1/getting-started-with-claude-cowork

How to evaluate tools before deploying →

Data shown here comes from public APIs and automated scanning. Reviewer notes reflect one person's experience. This is not a security certification or legal recommendation. Always evaluate tools according to your own organization's policies.

Evaluation

Ease of Use
4/5
Versatility
4/5
Reliability
4/5
Security
3/5
Overall score3.85 / 5.00 SolidEvaluatedMar 2026
Official Anthropic plugin. Connectors: Snowflake, Databricks, BigQuery, Definite, Hex, Amplitude, Jira. Works with plain CSV too — low barrier entry point. /data:explore is excellent for one-off analysis. Strong client-readiness because it democratises SQL.

← Back to Claude Tool Reviews