russelleNVy/three-man-team
A three-agent AI coding framework for software development using Claude Code with structured handoffs between Architect, Builder, and Reviewer roles.
Three Man Team is a structured multi-agent workflow system designed to improve AI coding assistance by assigning distinct roles to separate agents. The Architect plans and deploys, the Builder implements according to specification, and the Reviewer evaluates the output before acceptance. The system leverages Claude Code’s subagent capabilities to spin up secondary agents within a single session, with markdown-based context files (BUILDER.md, REVIEWER.md) managing handoffs and role definitions. Token optimization is a core design principle to reduce waste from redundant context and overly broad code retrieval.
Frequently asked
- What is russelleNVy/three-man-team?
- A three-agent AI coding framework for software development using Claude Code with structured handoffs between Architect, Builder, and Reviewer roles.
- Is three-man-team open source?
- Yes — russelleNVy/three-man-team is open source, released under the MIT license.
- What language is three-man-team written in?
- russelleNVy/three-man-team is primarily written in Shell.
- How popular is three-man-team?
- russelleNVy/three-man-team has 889 stars on GitHub.
- Where can I find three-man-team?
- russelleNVy/three-man-team is on GitHub at https://github.com/russelleNVy/three-man-team.