Use case

Researched articles

Articles drafted from real sources and checked until every claim is cited, then held for your approval before they’re saved as a Google Doc.

Researched articles: the workflow in the SynapseOrch editorResearched articles: the workflow in the SynapseOrch editor

The workflow in the editor, captured from the open-source app.

Steps
14
Agents
3
Where a person answers
2
Run side by side
1
Models
Yours to choose, per step: a small one for sorting and checks, a strong one where judgment matters
Stops at
Its turn and time limits; on Cloud, also your plan’s limit

The workflow

What it solves, and how it runs.

The problem

A single “write me an article” prompt produces confident text with no sources. Splitting it into research, drafting, review and revision produces something you can publish, and lets each step use a model priced for the job.

Why a fixed path matters here

Most of the work is research, and research doesn’t need your most expensive model. Put a small one on the gathering and a strong one on the writing.

How it runs

  1. 1A browsing agent gathers at least five sources
  2. 2A quality gate checks the research has sources and facts, and asks you for direction if not
  3. 3The draft is written while a second agent finds real-world examples, in parallel
  4. 4A review step checks every claim is cited, and sends the draft back for revision until it is
  5. 5You approve the article, or say what to change
  6. 6It’s saved as a Google Doc, and you get the link

Run it

On our cloud, or on your servers.

On SynapseOrch Cloud

The closest template:

  • Content pipelineBrief in, reviewed draft out: outline, write, critique, revise, approve.

Or describe this workflow to Build with AI, and approve the plan it drafts.

On your own servers

content-research-and-writing.bundle.json holds the agents and the workflow together. Import it from Settings → Import / Export in a self-hosted install, add your own connections, and run it.

It needs

  • Google Drive connected, for the document
  • A model for the agents, from any provider you set up
  • Web access: the built-in page reader, or a full browser when Node.js is installed

Two ways to run it

Let us run it, with nothing to install and no API key needed. Or run it on your own servers, free and open source.

Free plan, no card needed.