Connectors / Database
Database MCP connectors for Claude, ChatGPT and Cursor
10+ Database connectors in the catalog, each behind the same governed MCP endpoint as everything else in Elaichi.
https://api.elaichi.ai/mcp
The same for every user.
-
Their own access. An agent never gets more than the person it acts for.
-
One connection, every client. Claude, ChatGPT, Cursor, any MCP client, and the Elaichi Agent.
The connectors
Database connectors in the catalog
Connect one and your whole team reaches it, each inside the access they already have, without anyone handling a credential.
10 connectors
-
ClickHouse Cloud Hosted sign-in 13 tools
-
MotherDuck Hosted sign-in 39 tools
-
Neon Hosted sign-in 114 tools
-
Pinecone API key 47 tools
-
PlanetScale Hosted sign-in 25 tools
-
SAP HANA App credentials 12 tools
-
Sequin Stream API key 27 tools
-
Snowflake Bring your own OAuth app 205 tools
-
The Movie Database (TMDB) API key 151 tools
-
Upstash Hosted sign-in 55 tools
No Database connector by that name
It may sit under another category, or not be in the catalog yet — any documented API can become a connector.
In practice
What teams do with Database connectors
Drawn from the connector pages in this category, so every line describes something one of these connectors actually does.
Data
-
Match episodes to other catalogs
Ask for the external IDs of a TV episode in The Movie Database (TMDB) so it can be lined up against IMDb and other databases you already use. One question replaces a manual lookup across several sites.
-
Answer a question straight from the tables
Type the question in plain words, let the agent run the query against the right branch, and get the numbers back without opening a SQL editor.
-
See what a vector index holds
Ask how many vectors are in the product search index, what dimension it uses, and when it was last updated before you point a new feature at it.
Engineering
-
Check an index before a release
Ask how many vectors an index holds, which namespaces are filling up and whether the stats look right before shipping a change that depends on it.
-
Check a branch schema before merging
Ask for the schema on a PlanetScale branch, compare it with the main branch, and pull the open schema recommendations before a deploy request goes out.
-
See whether a ClickPipe is still flowing
List the ClickPipes in a service, open the one feeding the events table, and confirm it is healthy before a downstream team starts asking why the dashboard is stale.
Platform
-
Start a backfill after a schema change
Once a table has been altered, ask the agent to create a backfill for it in Sequin Stream and keep an eye on it until the consumer has caught up.
-
Find the queries slowing a project down
Pull the list of slow queries for a project, ask the agent to explain the worst one, and walk through query tuning before it reaches customers.
-
Check a Redis database before a launch
Ask for memory use, request counts and eviction settings on the database behind tomorrow's release, and get a plain answer without opening the console.
Analytics
-
Set up alerts on the numbers that matter
Create a Snowflake alert for a threshold your team watches, run it once to check the condition is right, then clone it for a second region.
-
Answer a metrics question without writing SQL
Ask for last quarter's revenue by region and the agent finds the right table in MotherDuck, runs the query and explains the result in plain language.
-
Check what a table really contains
Inspect the database, read the table schema and sample the rows before building a report on top of it, so the report starts from facts.
Once it is connected
Things to ask
Each of these is answered against the access the person asking already has, in the Database account you connected.
-
How many rows landed in the events table yesterday?
-
Which tables in the sales database were updated this week?
-
List the slow queries in the production project this week.
-
List all Pinecone indexes and their vector counts.
-
Show me the schema for the main branch of the orders database
-
List every SAP HANA user added in the last 30 days
-
List all Sequin Stream backfills started this week.
-
List Snowflake roles created in the last 30 days.
The tools
What an agent can call in Database
- Connectors
- 10+
- Tools
- 688
- Hosted sign-in
- 5
in the Database catalog
callable the moment you connect
connect with nothing to register
Of those 688 tools, 9% delete something. Restricting an agent to reads is not a promise here, it is 275 tools admitted and the rest left out — and a restricted tool is never advertised to the model at all.
- Read 40%
- 275 tools · list, get, search
- Write 16%
- 109 tools · create, update, send
- Delete 9%
- 62 tools · delete, remove, archive
- Other 35%
- 242 tools · vendor-specific verbs
One endpoint
https://api.elaichi.ai/mcp
Every connector above answers here.
Every Database connector, by depth
Tool counts are what the connector exposes today; the split is what those tools do.
Hosted sign-in takes a partnership with each vendor, and more are in progress. Until one lands, Your own app means the connector works today — you register an OAuth app once and connect. API key and App credentials mean an admin pastes them once, with nothing to register.
| Connector | Tools | Read · write · delete | Sign-in |
|---|---|---|---|
| Snowflake | 205 | Your own app | |
| The Movie Database (TMDB) | 151 | API key | |
| Neon | 114 | Hosted | |
| Upstash | 55 | Hosted | |
| Pinecone | 47 | API key | |
| MotherDuck | 39 | Hosted | |
| Sequin Stream | 27 | API key | |
| PlanetScale | 25 | Hosted | |
| ClickHouse Cloud | 13 | Hosted | |
| SAP HANA | 12 | App credentials |
FAQ
Frequently asked questions
How many Database connectors does Elaichi have?
10+ Database connectors are in the catalog today, and the list grows as connectors are added. Each one arrives as a set of MCP tools an agent can call through https://api.elaichi.ai/mcp.
Can Claude, ChatGPT and Cursor all use Database connectors?
Yes. Elaichi exposes one organization-wide endpoint, https://api.elaichi.ai/mcp, and Claude, ChatGPT, Cursor or any MCP client connects to that same address with OAuth, while the Elaichi Agent reaches the same tools inside the app. Connecting a Database account once makes it reachable from every one of them.
Do Database connectors work with Gemini, Codex, Claude Code or other MCP clients?
Yes. Database connectors are reached over the same MCP endpoint every client uses, so anything that speaks MCP can call them — Gemini, Codex, Claude Code, Windsurf, Cline, Zed and OpenCode among them — alongside Claude, ChatGPT, Cursor and the Elaichi Agent. There is no per-client setup beyond pointing the client at https://api.elaichi.ai/mcp.
Do Database connectors need me to bring my own OAuth app?
Most do not. 5 of the Database connectors use Elaichi's hosted sign-in, with nothing to register. 4 connect with an API key or app credentials that an admin pastes once, with no app to register. 1 asks you to bring your own OAuth app: you register it once with the vendor and connect. Hosted sign-in for those is in progress, a vendor partnership at a time, and when one lands the only thing that changes is that the registration step goes away. Each connector's page says which it is before you start.
Can I stop an agent from writing to Database tools?
Yes. Tool restrictions apply at role and individual level, and a restricted tool is left out of the model's tool list and cannot be called. Read-only access to a Database connector is a matter of allowing the reads and leaving the writes out.
Put Database connectors in front of your team
14 days on Gold, no credit card. Connect one and pick what each team can call.
- Works with
-
and any other MCP client, or the Elaichi Agent.
- When the trial ends
- Nothing is deleted. Connections, roles and the audit log stay where they are, so subscribing picks up exactly where you left off.