Sequel
MCP data layer that gives AI agents read-only access to marketing tools, billing data and warehouses with shared metric definitions.
Sequel is an MCP-based data layer for AI agents. Sequel connects tools such as GA4, Google Ads, HubSpot and Stripe, plus data warehouses, to Claude, ChatGPT, Cursor and other MCP clients with read-only access. Teams using Sequel connect sources, record their metric definitions and then ask questions from whichever agent they prefer.
Best for Marketing, revenue and finance teams that want AI assistants to answer from live company data using consistent metric definitions.
- Platforms
- MCP
- Pricing model
- Freemium
- Free plan
- Yes
- Made by
- Sequel
- Added
Last verified: . Information taken from sequel.sh.
What is Sequel?
Sequel is a data connection service that lets AI agents answer questions from a company's live marketing, revenue and product data. It runs over the Model Context Protocol, so one setup serves Claude, ChatGPT, Cursor, Codex, VS Code and other MCP clients. Sources include HubSpot, Stripe, Google Ads, Meta Ads, Google Analytics, Search Console, PostHog, Amplitude and Mixpanel, plus Postgres, BigQuery and Snowflake, linked through OAuth or a key. Teams record their own definitions for metrics such as qualified lead or blended CAC, and Sequel applies them to every answer. Connectors are read-only unless a write is explicitly enabled, credentials are held by Sequel and not exposed to the agent, and each answer comes with the query that produced it. Queries and actions are logged. A free plan exists, and an Enterprise option offers self-hosting.
Key Features
MCP Connection
Exposes connected data to any MCP-capable client, so switching between agents does not require reconnecting sources.
Source Connectors
Links analytics, ad platforms, CRM, billing, product analytics and warehouses through OAuth or an API key.
Metric Definitions
Stores how a team defines terms like active account or CAC and applies that definition consistently.
Read-Only by Default
Blocks writes unless enabled, and then requires the agent to show the exact call for confirmation before it runs.
Visible Queries
Returns the query behind each number so a result can be checked.
Audit Log
Records every query and action with attribution so activity can be reviewed.
How Sequel works
Connect your sources
Link analytics, ads, CRM, billing and warehouse accounts with OAuth or a key; nothing has to be self-hosted.
Teach it your definitions
State once what each metric means for your team so the same meaning is used every time.
Ask from your agent
Put questions to Claude, ChatGPT, Cursor, Codex or another MCP client and get answers with the query attached.
Screenshots
Use Cases
- Campaign Attribution
- Ask which campaign produced the most conversions last month and see the query used.
- Blended Acquisition Cost
- Calculate CAC per channel across ad platforms and CRM with one agreed definition.
- Pipeline Checks
- Have a revenue team query pipeline health from an AI assistant.
- Finance Snapshots
- Pull revenue, burn or margin figures from billing and warehouse data.
- Client Account Analysis
- Point an agent at a client's accounts with access limited to reading.
Things to know
- Writes are off unless enabled
- Connectors cannot modify campaigns, budgets or records until a write is switched on, and each write must be confirmed.
- Free plan limits
- The Free plan is capped at 3 data sources, 50 queries a month and a single user.
- Self-hosting is Enterprise only
- Self-hosted deployment, bring-your-own models and SSO with SAML appear only under the custom-priced Enterprise plan.
- Installed through a prompt
- The home page offers a prompt to paste into Claude Code, Codex or Cursor that installs Sequel and starts sign-in.
FAQs
What is Sequel?
Sequel is a data layer that connects marketing tools, billing systems and data warehouses to AI agents over the Model Context Protocol.
What is Sequel used for?
Sequel is used to ask AI assistants questions about live business data, such as campaign performance, CAC, revenue and signups.
Is Sequel free?
Yes, Sequel has a Free plan with 3 data sources, 50 queries per month and 1 user. No credit card is needed to connect a first source.
How much does Sequel cost?
Sequel: a free plan is available. Paid plan rates are published on sequel.sh. Check the official pricing page for current amounts.
Which AI tools does Sequel work with?
Sequel works with Claude, ChatGPT, Cursor, Codex, VS Code and any other client that supports MCP.
Which data sources does Sequel support?
Sequel supports ad platforms (Google Ads, Meta Ads), analytics tools (Google Analytics, Search Console, Amplitude, Mixpanel, PostHog), HubSpot, Stripe, Intercom, Google Sheets, and warehouses such as BigQuery, Snowflake and Postgres.
Can Sequel change data in my accounts?
Not by default. Sequel connectors are read-only, and a write runs only if you enable it and confirm the exact call.
Can Sequel be self-hosted?
Yes. Sequel lists self-hosted deployment, along with bring-your-own models and SSO, under its Enterprise plan.
Who it's for
- Marketing teams
- Revenue operations
- Finance teams
- Founders
- Agencies
- Data teams
Comments
No comments yet. If you have used Sequel, tell other readers how it went.
Alternatives to Sequel
| Tool | Free plan | Starting price | API |
|---|---|---|---|
 AirbyteOpen-source data movement platform that replicates data from 700+ sources and exposes it to analytics and AI agents. |
— | — | Yes |
 DatabricksUnified data and AI platform for data engineering, warehousing, analytics, governance and building AI agents. |
Yes | — | — |
 GongRevenue AI platform that records customer interactions and turns them into forecasting, coaching and sales automation. |
— | — | — |
 AjelixAgentic AI workspace that turns uploaded data into Excel files, dashboards, slide decks and other deliverables. |
Yes | — | — |
 PostHogProduct data platform combining analytics, session replay, feature flags, error tracking and a warehouse, with AI on top. |
Yes | — | Yes |
Listed in the same categories as Sequel. A starting price appears only where we checked it on the maker’s pricing page; a dash means not checked or not published. All Data Analysis & Analytics tools