Place single and batch voice AI calls, read call transcripts, and manage contacts, chats and the agent, phone-number, voice and knowledge-base catalogs on Synthflow AI.
Synthflow AI ships in the w6w first-party pack. It declares 26 actions, 2 health checks, and the host runs its code in a sandbox that never sees the credential.
io.w6w.synthflowSynthflow builds voice AI agents that answer and place phone calls. The unit this app moves around is the call: start an outbound call or a whole batch against an existing agent, watch a batch’s progress (pause, resume, cancel), and pull a finished call’s transcript and metadata. Around that sit the catalogs a workflow picks from — agents, phone numbers, voices, knowledge bases — plus contacts (create, update, delete) and text chat sessions.
Agent, action, knowledge-base, simulation and subaccount authoring is deliberately not covered: those are deeply nested configuration objects meant to be built once in the Synthflow dashboard and then referenced by id. A workspace lives in exactly one cluster (Global, US or EU) and a key only works there, so the connection asks for the region and the app reaches only that cluster’s host.
Health checks track the Synthflow API and End to End calling components of status.synthflow.ai alongside the derived credential check; Synthflow publishes no usage or rate-limit surface, so quota is declared unavailable.
Three routes to the same 26 actions. The Workflow tab is generated from Synthflow AI'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.
batch-call-create Create a batch of outbound calls, or append recipients to an existing batch.
batch-call-list-recipients List a batch's recipients, optionally by dispatch status.
batch-call-pause Suspend dialing; queued recipients stay queued. Allowed while scheduled or in progress.
call-list List calls placed or received by an agent, newest first, with optional filters.
chat-create Open a text chat session with an agent. The chat id is a UUID: supply one, or leave it blank to have one generated.
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 Synthflow AI's real ids.
{
"manifestVersion": "2",
"name": "synthflow-example",
"steps": [
{
"id": "assistant-get",
"uses": {
"app": "io.w6w.synthflow",
"action": "assistant-get",
"connection": "conn_YOUR_CONNECTION_ID"
},
"with": {
"model_id": "<model_id>"
}
}
]
}assistant-get assistant-list batch-call-create batch-call-get batch-call-list +21 more actions 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 Synthflow AI, 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 Synthflow AI. 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: "assistant-get",
payload: {
model_id: "<value>",
// include_actions: "<value>",
},
});
if (isActionRun(envelope)) console.log(envelope.value); npm install -g @w6w/cli w6w run conn_YOUR_CONNECTION_ID --action assistant-get --payload '{"model_id":"<value>"}' Give an AI agent Synthflow AI — without giving it Synthflow AI'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.synthflow#assistant-get",
"input": {
"model_id": "<model_id>"
}
}
}
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 Synthflow AI 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.
Synthflow AI'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. Synthflow AI itself is MIT, and the runtime that executes it is source-available (FSL).
Synthflow AI declares its own checks, so its health is a property of the app rather than something the host guesses at.
The `Synthflow API` and `End to End calling` components on status.synthflow.ai (incident.io).