Ask whether a Neon branch's compute is running, or why an operation failed.
Connect Neon with an API key and Memtro reads projects with region and Postgres version, branches with state, size and who created them, compute endpoints and whether they are active or idle, recent operations, databases on a branch, and usage history. It never returns connection strings, role passwords or the data in your tables. It reads on demand and there is no hourly sync. When asked, and once an administrator has granted the action by name, it can create, restore or permanently delete branches, start billable compute, suspend, restart or resize computes, take snapshots, create databases, and delete a project.
Ask in plain English from chat, your editor, a scheduled job or your own agent. The model picks the right Neon tool, reads what it needs, and answers with the source cited.
These tools change something in Neon. The model is told to use them only when you ask for exactly that, and an administrator grants each one by name to the teams that may use it; a personal connection acts only after its owner allows it.
Once Neon is connected, these are the kinds of things people ask. Plain questions, no syntax. If the answer lives partly in another system you have connected, Memtro looks there too.
Is the compute on the main branch active or idle?
Why did the database fail to start this morning?
Which branches were created this week, and by whom?
How many compute hours has the Acme project used this billing period?
Create a branch of main called invoice-fix with no compute
An engineer sees connection errors on the Northwind staging app and asks why. Memtro reads the project's recent operations, finds a failed compute start at 09:12 with its error, and confirms the endpoint has been active since 09:20.
The founder asks Memtro to delete four Neon preview branches that have not changed since August. Memtro names each back to her, and once she confirms deletes them permanently with their data, leaving the protected main branch and its children untouched.
The teams that lean on Neon most, a question each of them asks, and what they connect beside it.
A record in Neon is rarely the whole story. Connect the systems that hold the rest and Memtro checks them together, then remembers what it learned so the next question starts further along.
Using something else for the same job, or both? Each connector has its own tools; connect as many as you like.
Grist
NocoDB
Typesense
Weaviate
Tinybird
Hightouch
Census
kintone
Monte Carlo
OpenMetadata
Secoda
Softr
Teable
VectaraIn the Neon Console open Settings → API keys → Create new API key and copy it: Neon shows it once
A personal key works on any plan and reaches every organisation you belong to. An organisation or project-scoped key reaches less, but only an organisation admin can make one
In Memtro open Connections → Neon, paste the API key, add the Organization ID only if a personal key belongs to several, and connect. Usage history needs Launch or above
Ask whether the production branch's compute is awake
Chat in the dashboard, the OpenAI-compatible API, scheduled jobs, and any MCP client: Claude Code, OpenCode, Cursor, Claude Desktop. The Neon tools are the same everywhere.
Only when you ask, and only what it has been granted: create a branch, optionally with a compute (billed); rename a branch, protect or unprotect it, or set its expiry; make a branch the project's default; restore a branch to an earlier point in time (replaces its data); take a snapshot of a branch; delete a branch (permanent); create a database on a branch; start a compute (billed); suspend a compute (drops connections); restart a compute (drops connections); resize a compute or change its scale-to-zero timeout; delete a compute (drops connections); rename a project or change its history retention; delete a project (recoverable in Neon for 7 days). Every other tool reads, searches or lists. The Connections page labels the tools that act, an administrator grants each one by name to the teams that may use it, and a personal connection acts only after its owner allows it.
A personal connection serves only your own requests. A shared connection serves the organisation: everyone, or only the teams an admin chooses. Admins can always add shared connections; other members need the permission.
Credentials are encrypted at rest. Tool results are cached briefly so agent loops do not repeat identical calls, and the facts the model extracts are saved to your organisation's memory with provenance (system, record and link) so they can be checked and superseded.
API key. In the Neon Console open Settings → API keys → Create new API key and copy it: Neon shows it once.
Seven-day free trial, then one monthly price per organisation. Connectors, jobs and memory are never metered.
Sign in, add a provider key, create an API key and point your tools at Memtro. It takes about five minutes.
7-day free trial · card on file, nothing charged until it ends · cancel any time
