GitHub - coleam00/Archon: The first open-source harness builder for AI coding. Make AI coding deterministic and repeatable.
6 hours ago
- #open-source tools
- #workflow automation
- #AI coding
- Archon is an open-source harness builder for AI coding, designed to make AI coding deterministic and repeatable by defining development processes as YAML workflows.
- It acts as a workflow engine for AI coding agents, enabling reliable execution of tasks like planning, implementation, validation, code review, and PR creation across projects, similar to how Dockerfiles standardize infrastructure and GitHub Actions handle CI/CD.
- Key features include repeatability (consistent workflow sequences), isolation (git worktrees for parallel runs), fire-and-forget execution, composability (mixing deterministic and AI nodes), and portability (workflows work across CLI, Web UI, Slack, Telegram, and GitHub).
- Example workflows are provided for common development tasks, such as fixing GitHub issues, implementing features, and conducting PR reviews, with 17 default workflows included out of the box.
- Setup involves installing prerequisites (Bun, Claude Code, GitHub CLI), cloning the Archon repository, and using a setup wizard, with options for a full setup or a quick CLI install.
- Archon includes a web dashboard for chatting with coding agents, monitoring workflows, and building workflows visually, and supports integrations with platforms like Telegram, Slack, GitHub, and Discord.
- The architecture consists of platform adapters, an orchestrator for message routing, command handlers, workflow executors, AI assistant clients, and a database (SQLite/PostgreSQL) for managing codebases, conversations, and workflow runs.