Jev in production › SDKs and workflow integrations

AgentRun

Workflow DSL whose judge nodes send focused per-step questions to Jev and branch on answer and confidence, calling a full agent only when investigation is needed and escalating for review otherwise; ships as npm packages and a Pi extension.

Open on GitHub ↗

The source publishes no measured number.
Use
SDKs and workflow integrations
Industry
Developer tools
Form
Open-source tool
Stage
Beta
Plugs into
Pi
Listed
2026-09-24
Found via
github
Repository
Parcha-ai/agentrun
Stars
94
Forks
10
Last push
2026-10-11
Language
TypeScript
License
Apache-2.0
AgentRun screenshot
docs/assets/support-workflow.gif in the Parcha-ai/agentrun README, Apache-2.0; shown from GitHub.

The README opens with

AgentRun is a workflow language for the agents you already run. Define repeatable steps, use Jev for focused decisions, and call an agent when the work needs investigation. Your application keeps its tools, model access, permissions, and budgets.

Quickstart · Documentation · Examples · Pi extension · Agent instructions

The support example below follows this workflow with scripted tools and model responses.

This ticket-routing demo uses scripted decisions and needs no API key. Next, run and change a workflow in your own app. To run the support workflow shown above, use the checkout below. To build workflows with your agent, install the Pi extension.

Badge

For the project's own README, linking back here:

Listed in Jev in production

Also used for sdks and workflow integrations