GoKawiilParcha releases AgentRun, an open-source DSL for scripting AI agent workflows
Parcha has published AgentRun, a workflow language that lets developers define repeatable steps around existing AI agents, using a component called Jev for narrow decisions and invoking the agent itself only when deeper investigation is needed. The project is available via npm as a beta package, includes a demo ticket-routing workflow that runs without an API key, and ships with documentation, examples, and a Pi editor extension.
GoKawiil's interpretation of the reporting above, not reported fact.
By keeping an application's own tools, model access, permissions and budgets intact, AgentRun frames itself as a control layer rather than a replacement for existing agent stacks, which could appeal to teams wary of handing full autonomy to agents. The scripted demo, which requires no live API key, suggests the project is positioned for easy evaluation before developers commit to wiring in real models and tools.
- AgentRun is a new DSL from Parcha for building workflows around existing AI agents.
- It uses 'Jev' for scripted decision points and only calls the full agent for complex investigation tasks.
- A no-API-key demo and Pi extension are provided to let developers test and build workflows quickly.
Source: github.com, 2026-09-23
Published there as: “Show HN: AgentRun: DSL to turn agents into workflows”
Read the original report → The summary and analysis above are GoKawiil's own, written from reporting by the source above. Facts and quotes belong to the original publisher.