Open sourcing the coding agent harness I started an year ago
Summary
The author has open-sourced a self-hosted workspace for coding agents, featuring multi-agent architecture, AST-aware ChromaDB indexing, and a UI with Git diff management.
Similar Articles
I open-sourced the "harness" layer for AI agents: run Claude Code/Codex/Gemini with governed MCP tools (browser, editor, secrets)
Open-sourced a desktop workspace that provides a governed runtime for AI coding agents, offering 100+ MCP tools, RBAC, and a self-evolving toolbox.
Coding agents got good. Managing them didn't. I open-sourced the control room I built for it (MIT)
The author open-sourced o8, an MIT-licensed orchestrator that manages multiple coding agents in isolated git worktrees, with a merge gate, audit trail, and mobile approvals.
I open-sourced the agent instructions I use to keep my AI agents on track.
This open-source tool provides a portable, agent-agnostic control plane for coding agents, offering consistent operating rules, phased project planning, memory, skills, and runbooks via a set of files that can be dropped into any repository.
I built an open-source multi-agent SDLC harness that beats a cold Claude Code run on large repos, by learning the repo once. Real benchmarks (incl. where it loses) inside. [P]
AutoDev Studio is an open-source multi-agent AI coding harness that reduces costs by building a persistent repository knowledge base, outperforming cold-start Claude Code runs on large repos in most tasks.
Your agent is only as good as its harness. I open-sourced one with 40 capabilities behind a single function call
An open-source agent harness with 40 capabilities behind a single function call, including persistent memory, Docker sandbox, auto-summarization, stuck-loop detection, budget caps, and live run forking for branching agent execution. Built on Pydantic AI and designed to replace the 2000 lines of glue code every production agent needs.