Skip to content

runline

Code mode for agents: 204 bundled API plugins callable from sandboxed JavaScript, with a TypeSafe plugin exposing Jev as actions.

Pluginbeta

Overview

runline turns APIs and host capabilities into callable actions for agents. You install it, add connections for the plugins you want, and agents write JavaScript that runs in a QuickJS WASM runtime where each configured plugin is a global (for example brandfetch.brand.getColors or github.issue.create). It bundles 204 plugins for SaaS, DevOps, productivity and image-generation APIs, plus an optional node plugin for filesystem, shell and fetch access.

According to the list summary, a TypeSafe plugin exposes Jev decisions as callable actions, so agent code can ask a typed question and branch on the answer inside the same script that calls other APIs. The fetched README excerpt does not show the TypeSafe plugin itself.

What the decision model decides

TypeQuestionAnswers
OtherTyped Jev questions called as a plugin action from agent JavaScript(inferred)—

How it works

  1. runline init and add a TypeSafe connection
  2. Agent writes JavaScript calling plugin actions
  3. Call the TypeSafe action with a state and questions
  4. Branch on the answer and call other plugins

Strengths

  • Jev sits alongside 200+ API actions in one script
  • Sandboxed QuickJS runtime

Caveats

  • TypeSafe plugin not described in the fetched README
  • Licence not stated

Getting started

npm install -g runline && runline init, then add a connection and runline exec your script.

At a glance

Type
Plugin
Status
beta
Language
TypeScript
Runtime
CLI with QuickJS WASM runtime
Has code
yes

Use cases and integrations

How curated lists describe it

Frequently asked questions

How is Jev used in runline?

Per the list summary, a TypeSafe plugin exposes Jev decisions as actions callable from agent code.

Where does agent code run?

In a QuickJS WASM sandbox with configured plugins as globals.

How many plugins are bundled?

204, per the README.

Related projects

  • Reusable project context and specialist workflows for existing AI tools, with an addon that offloads bounded decisions to Jev.

    ProjectproductionTypeScript
  • Agent skill for writing and improving programs that call Jev: question design, state structure, thresholds and diagnosing wrong answers.

    Agent skillproduction
  • Go MCP server giving any client typed Jev decisions via OpenRouter or TypeSafe, with validated answers and no paid retries.

    MCP serverproductionGo
  • Open-source workspace for building, deploying and monitoring AI agents and workflows, listed as integrating TypeSafe System One.

    ProjectproductionTypeScript
  • DeepSeek Harness plugin that registers a native jev_decide tool and declares its own policy for when the agent should ask Jev.

    PluginbetaJavaScript
  • Agent Zero plugin adding a typesafe_query tool that asks Jev Choice, Noul and Score questions and shows probability cards in chat.

    ProjectbetaPython