Conductor is a Gemini CLI extension designed for Context-Driven Development, transforming the CLI into a proactive project manager. It enables developers to systematically specify, plan, and implement software features and bug fixes, ensuring a consistent, high-quality development lifecycle. By treating project context as a managed artifact, Conductor provides AI agents with deep, persistent project awareness, enhancing control and quality.

How It Works

Conductor enforces a strict Context -> Spec & Plan -> Implement protocol. It elevates project context (product goals, guidelines, tech stack, workflow) to a first-class artifact, stored alongside code. This context-driven approach ensures AI interactions are deeply informed by project specifics, leading to more coherent and controlled development. Plans are generated and reviewed before implementation, allowing for safe iteration and team-wide alignment on project foundations.

Quick Start & Requirements

• Primary install / run command: gemini extensions install https://github.com/gemini-cli-extensions/conductor --auto-update
• Non-default prerequisites and dependencies: Gemini CLI.
• Estimated setup time or resource footprint: Potential for increased token consumption, especially in larger projects or during extensive planning/implementation phases.
• Links: Documentation for Gemini CLI extensions, GitHub issues for bug reports/feature requests.

Highlighted Details

• Plan Before Build: Generates detailed specs and actionable plans before code generation.
• Context Management: Maintains AI adherence to style guides, tech stack, and product goals.
• Iterative Development: Allows plan review before implementation, enabling safe iteration.
• Team Collaboration: Establishes shared project context for team consistency.
• Smart Revert: Git-aware reversion of logical work units (tracks, phases, tasks).
• Project Initialization: Supports both new (Greenfield) and existing (Brownfield) projects.

Maintenance & Community

The project relies on GitHub Issues for bug reporting and feature requests. No specific community channels (e.g., Discord, Slack) or details on core contributors/sponsorships are provided in the README.

Licensing & Compatibility

• License: Apache License 2.0.
• Compatibility: Permissive license suitable for most commercial use cases.

Limitations & Caveats

A primary caveat is potential for increased LLM token consumption due to the extensive context analysis required for its context-driven development approach, especially in large or complex projects.