Andrii Bidochko

• Updated: February 2, 2026
• 3 min read

Google Conductor: A Context‑Driven Gemini CLI Extension Revolutionizing AI Agent Development

Google Conductor: A Context‑Driven Gemini CLI Extension Revolutionizing AI Agent Development

Google has unveiled Conductor, an open‑source extension for the Gemini Command‑Line Interface (CLI) that brings a new level of context‑driven development to AI agents. By storing project knowledge as version‑controlled Markdown and orchestrating agentic workflows, Conductor aims to make AI‑powered software development more repeatable, auditable, and collaborative.

Key Features

• Versioned Markdown Context: All project specifications, plans, and notes are kept in Markdown files that are automatically versioned, giving a clear audit trail.
• Track‑Based Workflow: Projects are organized into tracks, each containing a spec.md (what to build) and a plan.md (how to build it).
• CLI Commands: Simple commands such as conductor init, conductor implement, conductor status, conductor review, and conductor rollback guide developers through the entire lifecycle.
• Agentic Orchestration: The extension can invoke Gemini agents to generate code, run tests, and produce documentation based on the stored context.

Installation & Quick Start

1. Ensure you have the Gemini CLI installed (v0.5.0 or later).
2. Run npm i -g @google/conductor to install the extension globally.
3. Initialize a new project with conductor init my‑project. This creates a .conductor folder with the required structure.
4. Edit spec.md to describe the desired AI agent behavior and plan.md for step‑by‑step implementation details.
5. Execute conductor implement to let Gemini generate the code based on your specifications.

How the Workflow Works

Conductor treats the spec.md and plan.md files as the single source of truth. When you run conductor implement, the CLI reads the Markdown, sends it to the Gemini model, and writes the generated artifacts back into the project directory. Subsequent commands let you check status, review changes, or rollback to a previous version, all while preserving the full history in Git.

Impact on AI Agent Development

By coupling Markdown‑based documentation with automated code generation, Conductor bridges the gap between high‑level design and implementation. Teams can now:

• Maintain a clear, searchable knowledge base for each AI agent.
• Iterate rapidly while keeping a full audit trail of decisions.
• Share reproducible workflows across organizations, reducing onboarding time.

Community and Resources

The project is open‑source and hosted on GitHub, inviting contributions from the broader AI community. For a deeper dive, read the original announcement on MarkTechPost.

Explore related UBOS resources to extend your AI workflows:

• AI Agents Overview
• Gemini CLI Guide
• Markdown‑Based Development Workflows

Call to Action

Ready to experiment with context‑driven AI development? Visit the Conductor repository to get started, and check out UBOS’s tutorials to integrate it seamlessly into your projects.

Andrii Bidochko

CTO UBOS

Andrii Bidochko is an AI entrepreneur and researcher focused on AI agents, reinforcement learning, and autonomous systems. He writes about the technologies shaping the future of machine intelligence, from frontier models and agent architectures to real-world AI applications.