Open Orchestrators Multi-agent orchestration profiles
Back to directory
Ag

Player profile

Agon

An autonomous research system that coordinates scientist, coder, and auditor loops from topic to idea, proposal, experiment, and paper.

GitHub Paper

Profile

Where it fits

Agon is an autonomous research system for running scientific workflows from a topic through idea generation, proposal writing, experiment execution, and paper drafting.

It belongs in Open Orchestrators because the repository is organized around multi-agent research loops rather than a single assistant: scientist, coder, and auditor roles are dispatched through reusable prompts, with critics used to break outputs before the workflow moves forward.

The public repository includes setup instructions and agent, command, reference, and skill artifacts. The accompanying paper frames the design around Prompt Economy and zero-code orchestration.

Builder web analytics

Measure projects built with Agon

Agon can run long research loops that produce project pages, papers, demos, docs, and benchmark artifacts. Agent Analytics gives the human operator a way to measure whether those public surfaces reached readers and users after release.

Instrument the public surface created around the Agon output. Agent Analytics measures user-facing web and product events; it does not replace Agon's internal auditor reports, run logs, or human scientific judgment.

First loop to measure

  1. an Agon run produces or updates a research page, docs path, demo, benchmark artifact, or paper companion site
  2. the released surface reports visits, sources, CTA clicks, downloads, signups, paper clicks, or benchmark interactions to Agent Analytics
  3. a follow-up agent fetches the Agent Analytics results after the release
  4. the operator uses the measured outcome to decide which research artifact, demo, or explanation to improve next

Copyable prompt

Use Agent Analytics for this project. If event reporting is missing, add the tracker and report events for this project surface, including Agon-built research page, docs path, paper link, demo, benchmark artifact, traffic source, or release outcome. Verify events are arriving. Then fetch the last 7 days and compare them with the prior 7 days. Tell me which Agon-built research page, docs path, paper link, demo, benchmark artifact, traffic source, or release outcome moved users toward value, where users dropped off, which sources mattered, and what my agent workflow should improve next.
Track builder outcomes

Related

More on Agon