coleam00/Archon

The first open-source harness builder for AI coding. Make AI coding deterministic and repeatable.

What it solves

AI coding agents often produce inconsistent results because they may skip planning, forget to run tests, or ignore project templates. Archon provides a way to make AI-driven software development deterministic and repeatable by encoding development processes into structured workflows.

How it works

Archon is a workflow engine that allows users to define development processes as YAML files. These workflows consist of nodes that can be AI-powered (for planning, code generation, and review) or deterministic (such as bash scripts, tests, and git operations). The engine manages the execution sequence, including loops for iterative implementation and human-approval gates to pause for feedback.

To ensure isolation, every workflow run is executed in its own git worktree, preventing conflicts when running multiple tasks in parallel. It integrates with AI assistants like Claude Code and can be triggered via a CLI, a Web UI, or platform adapters for Slack, Telegram, and GitHub.

Who it’s for

Software developers who want to automate complex coding tasks—such as fixing GitHub issues or implementing features from a PRD—while maintaining strict control over the process and structure.

Highlights

  • Deterministic Workflows: Define exact sequences of planning, implementation, and validation in YAML.
  • Worktree Isolation: Each run gets a dedicated git worktree for parallel, conflict-free execution.
  • Hybrid Nodes: Mixes AI intelligence with deterministic bash scripts and tests.
  • Multi-Platform Access: Operates via CLI, Web UI, and integrations with Slack, Telegram, and Discord.
  • Extensive Default Library: Includes 19 pre-built workflows for tasks like adversarial development, safe refactoring, and PR validation.

Related

  • Project
    microsoft/conductor

    A CLI tool for defining and running deterministic, version-controlled multi-agent workflows using YAML configurations and the GitHub Copilot SDK.

  • Project
    nrslib/takt

    A CLI tool that orchestrates AI coding agents into repeatable, YAML-defined development workflows with isolated worktrees and explicit review loops.

  • Project
    aeonfun/aeon

    An autonomous agent framework that runs unattended on GitHub Actions to ship code, deploy apps, and find vulnerabilities using a prompt-based skill system.

  • Project
    google/adk-python

    A code-first Python framework for building, evaluating, and deploying sophisticated AI agents using graph-based workflows and multi-agent orchestration.

  • Project
    SynkraAI/aiox-core

    An open-source AI orchestration framework that implements an agentic agile workflow to prevent context loss and planning inconsistency in AI-assisted software development.