Database · Integration

AI agents for PostgreSQL

The production database your reports should come from. AGNT connects to PostgreSQL so agents can read what happens there, reason about it, and act — with a receipt for every run.

Install Report Assembler → Get AGNT free

Report Assembler is a ready-made agent that works in PostgreSQL — free, and it runs on your machine.

PostgreSQL and AGNT, in brief

What it does	Lets an AGNT agent read from and act in PostgreSQL.
Connects with	Connection string — held in the local vault
AGNT node	custom-api, or the zapier-action bridge
Runs when	A scheduled query runs
Typical build time	Ten minutes
Where data goes	Nowhere by default — AGNT runs on your machine and credentials never leave it.

What you can automate

Reacts when…

a scheduled query runs
a threshold is crossed

Agents can…

run parameterised queries
write validated rows
produce report extracts

Teams use it for

metrics pulled straight from source
anomalies caught by scheduled checks
clean extracts for the weekly report

Build it: the PostgreSQL workflow

Take the simplest useful version: a scheduled query runs in PostgreSQL. Instead of a rule firing blindly, an agent reads the item in full, works out whether it matters and what it means, and then decides whether to run parameterised queries. Over a week that turns metrics pulled straight from source from something a person checks into something that reports to a person.

The shape is deliberately small. A trigger watching PostgreSQL, an agent that judges what arrived, and one action taken only when that judgment warrants it. Everything below is the actual definition, not an illustration of one.

The whole flow runs locally against your own PostgreSQL account. Deciding to take no action counts as an outcome and is recorded like any other.

PostgreSQL — read, decide, actpaste into a new workflow

{
  "id": "d4eb8cc1-c598-4426-a22d-ce4703dad880",
  "name": "PostgreSQL — read, decide, act",
  "nodes": [
    {
      "id": "8ac57a6b-420e-4f6e-a45b-485d1a343534",
      "text": "Run on a schedule",
      "x": 512,
      "y": 144,
      "isEditing": false,
      "type": "trigger-timer",
      "icon": "clock",
      "category": "trigger",
      "isSelected": false,
      "parameters": {
        "interval": "15",
        "unit": "minutes"
      },
      "description": "Checks PostgreSQL every fifteen minutes.",
      "error": null,
      "isActive": false,
      "output": null,
      "outputs": {}
    },
    {
      "id": "1f042846-83ea-4af6-ad72-f2ba97321280",
      "text": "Work out what should change",
      "x": 512,
      "y": 336,
      "isEditing": false,
      "type": "agnt-agent",
      "icon": "agnt",
      "category": "action",
      "isSelected": false,
      "parameters": {
        "instructions": "Decide what needs to happen in PostgreSQL. Explain your reasoning."
      },
      "description": "Decides what needs to happen in PostgreSQL, and why.",
      "error": null,
      "isActive": false,
      "output": null,
      "outputs": {}
    },
    {
      "id": "a40dd042-0792-4ee3-a568-31e96c1df342",
      "text": "Call the PostgreSQL API",
      "x": 512,
      "y": 528,
      "isEditing": false,
      "type": "custom-api",
      "icon": "connect",
      "category": "action",
      "isSelected": false,
      "parameters": {
        "url": "postgresql://{user}:{password}@{host}:5432/{database}",
        "method": "SQL",
        "authType": "Bearer",
        "headers": "{\"Content-Type\": \"application/json\"}"
      },
      "description": "SQL against the PostgreSQL API, authenticated with connection string.",
      "error": null,
      "isActive": false,
      "output": null,
      "outputs": {}
    }
  ],
  "edges": [
    {
      "id": "a4aba3e1-f795-4108-a617-1df78f29c3f0",
      "start": {
        "id": "8ac57a6b-420e-4f6e-a45b-485d1a343534",
        "type": "output"
      },
      "end": {
        "id": "1f042846-83ea-4af6-ad72-f2ba97321280",
        "type": "input"
      },
      "startX": 800,
      "startY": 168,
      "endX": 512,
      "endY": 360
    },
    {
      "id": "1160527d-8d60-4563-a481-9396a02346ca",
      "start": {
        "id": "1f042846-83ea-4af6-ad72-f2ba97321280",
        "type": "output"
      },
      "end": {
        "id": "a40dd042-0792-4ee3-a568-31e96c1df342",
        "type": "input"
      },
      "startX": 800,
      "startY": 360,
      "endX": 512,
      "endY": 552
    }
  ],
  "zoomLevel": 1,
  "canvasOffsetX": 0,
  "canvasOffsetY": 0,
  "isTinyNodeMode": false
}

Install Report Assembler — it runs on PostgreSQL

Report Assembler is a working agent that uses PostgreSQL. Installing it drops the whole workflow onto your canvas — connected, editable, and yours to change.

Install Report Assembler → Don’t have AGNT? Get it free

Runs on your own machine · See what it does

Reports should come from the source

Numbers copied into a spreadsheet are stale the moment they are pasted. Querying the production database at run time means a report describes the system as it is, and the query is written once rather than reconstructed each cycle.

Parameterised, not improvised

Database access from an agent needs the same discipline as from any application: parameterised queries, a scoped role, and read-only credentials wherever writing is not required. The scoping belongs to the connection rather than to the model’s good behaviour.

PostgreSQL without a first-party node

There is no first-party PostgreSQL node yet, and a page implying otherwise would cost you an afternoon. What works today is the generic custom-api node pointed at PostgreSQL’s own API, authenticating with connection string.

What you need	Value
Endpoint	`postgresql://{user}:{password}@{host}:5432/{database}`
Method	`SQL`
Authentication	Connection string
Header	`n/a — credentials live in the connection string`
Reference	PostgreSQL API documentation

Worth knowing: Not an HTTP API. Reach it from an execute-javascript node, or put PostgREST in front of it if you want custom-api to work directly.

Where PostgreSQL integrations usually go wrong

Not an HTTP API. Reach it from an execute-javascript node, or put PostgREST in front of it if you want custom-api to work directly.

It is worth getting right first, because the failure is quiet rather than loud — you get a response, it just is not the one you wanted. Once past it, PostgreSQL is a straightforward API to automate against.

Connect PostgreSQL in two minutes

Download AGNT Community Core — free, local-first, no account needed to run.
Connects to your local instance directly — no cloud account required, nothing leaves your machine.
Drop the PostgreSQL node into a workflow or hand it to an agent — the first run produces a receipt you can read line by line.