Read and write records in an Adalo app's Collections via the Adalo Collections API.
Adalo ships in the w6w first-party pack. It declares 5 actions, 2 health checks, and the host runs its code in a sandbox that never sees the credential.
io.w6w.adaloAdalo is a no-code app builder whose apps each carry their own built-in database, organized into Collections (roughly: tables) of Records (rows). This app talks to the Collections API that every Adalo app on the Team or Business plan exposes at a fixed host — listing, reading, creating, updating and deleting records inside a named Collection.
Because every Adalo app defines its own Collections and fields, the shapes here stay generic: a Collection ID and a JSON object of field slug → value, rather than a fixed field list, so this app works against any Adalo application without per-schema code. Adalo does not publish an endpoint that enumerates an app’s Collections or their field schemas — those are only visible from the app’s own generated API Documentation panel in the Adalo builder (Collection menu → API Documentation) — so a workflow author copies the App ID, Collection ID and API key from there once and supplies the Collection ID on every action.
Left out on purpose: listing Collections (no such endpoint exists) and the separate Notifications API (a differently-shaped, non-collection endpoint for pushing a notification to one user, out of scope for this Collections-focused app).
Three routes to the same 5 actions. The Workflow tab is generated from Adalo'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.
list-records List the records in an Adalo Collection, optionally filtered by one property.
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 Adalo's real ids.
{
"manifestVersion": "2",
"name": "adalo-example",
"steps": [
{
"id": "create-record",
"uses": {
"app": "io.w6w.adalo",
"action": "create-record",
"connection": "conn_YOUR_CONNECTION_ID"
},
"with": {
"collectionId": "<collectionId>",
"fields": "<fields>"
}
}
]
}create-record get-record list-records update-record +1 more action available
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 Adalo, 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 Adalo. 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: "create-record",
payload: {
collectionId: "<value>",
fields: "<value>",
},
});
if (isActionRun(envelope)) console.log(envelope.value); npm install -g @w6w/cli w6w run conn_YOUR_CONNECTION_ID --action create-record --payload '{"collectionId":"<value>","fields":"<value>"}' Give an AI agent Adalo — without giving it Adalo'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.adalo#create-record",
"input": {
"collectionId": "<collectionId>",
"fields": "<fields>"
}
}
}
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 Adalo connections to sign
with, and refuses rather than guesses when the answer is ambiguous.
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.
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.
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.
Adalo'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. Adalo itself is MIT, and the runtime that executes it is source-available (FSL).
Adalo declares its own checks, so its health is a property of the app rather than something the host guesses at.
status.adalo.com component tree, scoped to the 'Collections API' component. Unauthenticated and unsigned.