The datasets in your BigQuery project, the tables and views inside each one, and every column’s name and type — answered inside ChatGPT, Claude, Codex or Cursor, so the person writing the query starts from the real schema. Connect BigQuery to Soku once; your AI reads it with your own credentials, and every read costs 0 credits.
Set it up in
Sign in, connect the account once, paste the setup prompt to your agent, ask something. That’s the whole setup.
Your accounts, one connection
ads, analytics, commerce and revenue
BigQuery
Datasets, tables and schemas in your warehouse
AppsFlyer
Installs, ROAS and retention by media source
PostHog
Product events, funnels and retention
200+
APIs
40+
apps
30+
skills
Ask once and your AI reads the accounts you connected — no paid data call involved.
Connect your BigQuery account in Soku once. After that, every AI app you approve reads it through the same connection, with nothing to install for BigQuery itself.
Claude Code
One sign-in covers every project, and headless runs (`claude -p`, CI, the Agent SDK) work without a browser — `claude mcp add` only writes config, so a non-interactive run can never finish the MCP authorization. Both surfaces can act on your accounts once you approve each change.
Create your Soku workspace
The connection is public and identical for everyone — your Soku sign-in is what points it at your data. Connect one account in that workspace too: that is what your AI reads. The first time your AI uses the connection, Soku asks you to approve the access and names the workspace it will read through — so the workspace has to exist by then.
Already have one? Nothing to do here — sign in when the connection asks for it.
Copy the Soku MCP connection
https://api.soku.ai/mcpAdd it to Claude Code
claude mcp add --transport http soku https://api.soku.ai/mcp
Approve the access, then ask it something
Paste this first
Using Soku, search for what capabilities I can reach.
A list of capabilities with names, what each reads and what each costs. An empty list means the search matched nothing — try broader words. An error naming authorization means the host never completed the browser step.
Checked 2026-08-25 against Claude Code documentation. Menu names change between releases — look for connectors, apps, tools or integrations.
Each one is a tool your AI is offered by name once you connect. You never have to type the name — it is here so you can tell what actually ran.
bigquery_list_datasetsEvery dataset the connection can see, with its region — the starting point for knowing what data exists at all.
bigquery_list_tablesTables, views, materialised views and external tables in one dataset, each with its type — views are often the curated entry point, and they are listed alongside tables.
bigquery_get_table_schemaA table or view’s columns — name, type and mode, with nested fields one level deep — plus its row and byte counts.
Start free, upgrade when your team needs more credits, more brands, or deeper workflow execution.
Free
$0/month
Free forever
2,000 credits to start
For trying Soku — no credit card required.
Creator
$15/month
Billed $180/year· Save $48/yr
84,000 credits / year
For one person shipping creative on a budget.
Starter
$39/month
Billed $468/year· Save $120/yr
240,000 credits / year
For individual creators and solo operators.
Pro
$199/month
Billed $2388/year· Save $600/yr
2,400,000 credits / year
For teams with frequent AI workflow execution.
Enterprise
Custom
Flexible contract
Unlimited credits
For organizations requiring scale, security, and control.
Prices in USD. Cancel anytime. Annual plans are billed yearly at a 20% discount.