Vibestrate

Vibestrate is an open-source orchestration harness designed for managing structured AI coding workflows and autonomous execution pipelines, maintained by developer guyshonshon. It provides an orchestration layer that standardizes multi-step development flows, step boundaries, and task verification across repositories. Day-to-day interaction takes place primarily through the `vibe` command-line interface paired with a local browser-based dashboard built with React, xterm, and CodeMirror. Vibestrate manages execution runs through configurable flow schemas, tracking intermediate steps, diffs, and execution status while offering built-in terminal documentation browsing. Unlike standard single-turn coding agents, Vibestrate focuses on flow governance, execution observability, and scope control. It features an architect scope contract that constrains agent task scopes and introduces merge gating alongside interactive commands like `vibe steer` to dynamically adjust execution state during complex tasks. Vibestrate is distributed as an open-source package via npm with full bundle-level dependency licensing attribution.

Tags: agent harness, cli, code assistant, coding agent, developer tools, OpenSource

Visit Vibestrate

Key features

• Flow-based orchestration — Coordinates multi-step AI coding runs using structured schemas and derived flow pipelines. - Architect scope contract — Constrains agent modifications within declared architectural boundaries prior to merge gates. - Runtime steering — Supports live execution intervention and dynamic task guidance via commands like vibe steer. - Web inspection dashboard — Bundles a local visual interface with xterm terminal playback, visual graph views, and CodeMirror diff viewing. - In-shell documentation — Includes a built-in terminal markdown viewer for checking flow schemas and command flags offline. - Merge gate verification — Validates agent-generated code changes against defined pipeline criteria before marking tasks ready.

Use cases

Use Case: Enforcing Scope Limits on AI Refactoring Developers running multi-file or multi-step agent tasks can define architect scope contracts to prevent agents from modifying out-of-scope files before reaching a merge-ready state.

Use Case: Live Steering of Agent Pipelines When an automated run begins deviating from expected implementation patterns, engineers can monitor the process and use vibe steer to guide intermediate steps without restarting the entire run.

Use Case: Visual Observability for Orchestrated Runs Teams running complex pipelines can launch the local web dashboard to inspect step logs, xterm execution output, and visual flow graphs across different task phases.

Who it is for

Vibestrate is suited for software engineers and teams that need structured pipeline orchestration, strict architectural scope boundaries, and execution monitoring over AI coding tasks. It is not intended for developers looking for a lightweight, single-turn chat CLI without workflow orchestration.

Install guide

Installation Install Vibestrate globally using npm: bash npm install -g vibestrate

First Run Verify the installation and check the integrated documentation: bash vibe --help

Example Session Interact with or steer an active orchestration workflow: bash # Inspect and steer an active flow execution vibe steer

Part of CLI Coding Agents

Categories: AI Coding Agents, Code Assistants, Developer Tools

Related CLI agents

• Vibe Kanban — Vibe Kanban, originally developed by Bloop AI, is an open-source orchestration harness and visual task board designed to manage and…
• Claudexor — Claudexor is an open-source agent harness and multi-agent orchestrator created by razzant. Built as a monorepo project, it provides an…
• Parallel Code — Parallel Code is an open-source agent orchestration platform and desktop harness developed by Johannes Millan (johannesjo). Built on an Electron…
• Cline — Cline is an open-source autonomous coding agent runtime maintained by Cline Bot Inc. and an active open-source community. Available as…
• Continue — Continue is an open-source AI code assistant and agent framework originally developed by the Continue team (continuedev) and recently acquired…
• SWE-agent — SWE-agent is an open-source autonomous software engineering agent harness built and maintained by researchers from Princeton University and Stanford University.…