PostgreSQL
Run SQL directly against a PostgreSQL database over its own wire protocol.
About
PostgreSQL ships in the w6w first-party pack. It declares 2 actions, 1 health check, and the host runs its code in a sandbox that never sees the credential.
- App id
io.w6w.postgres- Version
- 0.1.1
- Author
- w6w
- Licence
- MIT
- Categories
- Databases
Overview
PostgreSQL is a raw database, not an HTTP API — there is no vendor REST surface to call, so this app speaks the Postgres wire protocol (v3.0) directly: it authenticates with SCRAM-SHA-256, the mechanism every current PostgreSQL install negotiates by default, and sends a workflow’s SQL to the server exactly as written.
The connect form collects host, port, database, TLS mode, and a username/password pair once; the host resolves and pre-checks that target before ever dialing it, and the SCRAM exchange runs as an iterative, network-less hook so the password itself never leaves the sandbox that holds it. query then runs over the resulting authenticated socket and returns whatever rows come back.
There is deliberately no parameter-binding syntax ($1, a params array) in this version — the caller’s SQL is sent verbatim as a Simple Query, so building a statement safely is the caller’s job, the same way it would be with psql or any other client that has not yet added prepared statements.
Build with PostgreSQL
Three routes to the same 2 actions. The Workflow tab is generated from PostgreSQL's own manifest and carries its real ids, so it is copy-pasteable; the Code and CLI examples are the same call for any action on any app, so every app-specific value in them is a blank you fill in.
A workflow step names the app and the action, and the editor fills in the
connection when you pick one. This is the Step shape from the
workflow spec, carrying PostgreSQL's real ids.
{
"manifestVersion": "2",
"name": "postgres-example",
"steps": [
{
"id": "execute",
"uses": {
"app": "io.w6w.postgres",
"action": "execute",
"connection": "conn_YOUR_CONNECTION_ID"
},
"with": {
"sql": "<sql>"
}
}
]
}Here are some of the things you can do
-
Execute
performexecute -
Query
performquery
Every app-specific value here is a blank you have to fill in. An
app action is reached through the connection that authenticates it, so the
address is a connection id, not the app id — and connections belong to your account,
so a public page cannot know yours. Create one for PostgreSQL, then fill in
the three blanks: conn_YOUR_CONNECTION_ID, the action key, and the
parameters that action declares. The call itself is real — the shape is transcribed
from the studio's own snippet builder, which prints the same kind of blanks — but
nothing in it is specific to PostgreSQL. The Workflow tab is where this app's
real ids are.
npm install @w6w/sdkyarn add @w6w/sdkpnpm add @w6w/sdkdeno add npm:@w6w/sdkimport { W6wClient, isActionRun } from "@w6w/sdk";
// Reads W6W_BASE_URL and W6W_TOKEN from the environment when omitted.
const client = new W6wClient();
const envelope = await client.run({
urn: "conn_YOUR_CONNECTION_ID",
action: "execute",
payload: {
sql: "<value>",
},
});
if (isActionRun(envelope)) console.log(envelope.value);npm install -g @w6w/cliw6w run conn_YOUR_CONNECTION_ID --action execute --payload '{"sql":"<value>"}'Give an AI agent PostgreSQL — without giving it PostgreSQL's credentials. One MCP endpoint exposes every app, function and workflow the caller is entitled to, as tools it can discover and run. Access is granted per team while we onboard.
One tool call{
"name": "w6w_invoke",
"arguments": {
"ref": "app:io.w6w.postgres#execute",
"input": {
"sql": "<sql>"
}
}
}
Every tool names its target with a single ref. The
app: form above doesn't name a connection at all — the
host resolves which of the caller's PostgreSQL connections to sign
with, and refuses rather than guesses when the answer is ambiguous.
What the agent gets
Credentials it can't read
The token is attached host-side, at the moment of the call. It is never a tool argument, never in the model's context, and never in a transcript — so a prompt injection has nothing to exfiltrate.
A tool surface scoped to the caller
Tools are derived per end user from what that person has actually connected and is entitled to — not one shared bot identity carrying the union of everyone's access.
A durable workflow in one call
Multi-step work runs on the workflow engine and returns a run handle the agent can poll — retries, branching and state survive the conversation that started them.
Health-aware discovery
PostgreSQL's declared health checks are on the surface too, so an agent can tell "the vendor is down" from "your credential expired" before it burns a retry on either.
The MCP surface is part of the hosted platform. PostgreSQL itself is MIT, and the runtime that executes it is source-available (FSL).
Health checks
PostgreSQL declares its own checks, so its health is a property of the app rather than something the host guesses at.
Database reachable
Runs SELECT 1 over an authenticated connection to confirm the database actually answers queries, not just that the stored credential looks well-formed.