{"slug": "open-source-harness-builder-for-ai-coding", "title": "Open-source harness builder for AI coding", "summary": "Archon, an open-source workflow engine for AI coding agents, has been released to make AI coding deterministic and repeatable by defining development processes as YAML workflows. The tool, created by coleam00, allows developers to encode phases like planning, implementation, validation, and PR creation, with the AI filling in intelligence at each step while the structure remains deterministic. Archon is positioned as a Dockerfile or GitHub Actions equivalent for AI coding workflows, offering repeatability, isolation via git worktrees, fire-and-forget execution, composability of deterministic and AI nodes, and portability across CLI, Web UI, Slack, Telegram, or GitHub.", "body_md": "The first open-source harness builder for AI coding. Make AI coding deterministic and repeatable.\n\nArchon is a workflow engine for AI coding agents. Define your development processes as YAML workflows - planning, implementation, validation, code review, PR creation - and run them reliably across all your projects.\n\nLike what Dockerfiles did for infrastructure and GitHub Actions did for CI/CD - Archon does for AI coding workflows. Think n8n, but for software development.\n\nWhen you ask an AI agent to \"fix this bug\", what happens depends on the model's mood. It might skip planning. It might forget to run tests. It might write a PR description that ignores your template. Every run is different.\n\nArchon fixes this. Encode your development process as a workflow. The workflow defines the phases, validation gates, and artifacts. The AI fills in the intelligence at each step, but the structure is deterministic and owned by you.\n\n**Repeatable**- Same workflow, same sequence, every time. Plan, implement, validate, review, PR.** Isolated**- Every workflow run gets its own git worktree. Run 5 fixes in parallel with no conflicts.** Fire and forget**- Kick off a workflow, go do other work. Come back to a finished PR with review comments.** Composable**- Mix deterministic nodes (bash scripts, tests, git ops) with AI nodes (planning, code generation, review). The AI only runs where it adds value.**Portable**- Define workflows once in`.archon/workflows/`\n\n, commit them to your repo. They work the same from CLI, Web UI, Slack, Telegram, or GitHub.\n\nHere's an example of an Archon workflow that plans, implements in a loop until tests pass, gets your approval, then creates the PR:\n\n```\n# .archon/workflows/build-feature.yaml\nnodes:\n  - id: plan\n    prompt: \"Explore the codebase and create an implementation plan\"\n\n  - id: implement\n    depends_on: [plan]\n    loop:                                      # AI loop - iterate until done\n      prompt: \"Read the plan. Implement the next task. Run validation.\"\n      until: ALL_TASKS_COMPLETE\n      fresh_context: true                      # Fresh session each iteration\n\n  - id: run-tests\n    depends_on: [implement]\n    bash: \"bun run validate\"                   # Deterministic - no AI\n\n  - id: review\n    depends_on: [run-tests]\n    prompt: \"Review all changes against the plan. Fix any issues.\"\n\n  - id: approve\n    depends_on: [review]\n    loop:                                      # Human approval gate\n      prompt: \"Present the changes for review. Address any feedback.\"\n      until: APPROVED\n      interactive: true                        # Pauses and waits for human input\n\n  - id: create-pr\n    depends_on: [approve]\n    prompt: \"Push changes and create a pull request\"\n```\n\nTell your coding agent what you want, and Archon handles the rest:\n\n```\nYou: Use archon to add dark mode to the settings page\n\nAgent: I'll run the archon-idea-to-pr workflow for this.\n       → Creating isolated worktree on branch archon/task-dark-mode...\n       → Planning...\n       → Implementing (task 1/4)...\n       → Implementing (task 2/4)...\n       → Tests failing - iterating...\n       → Tests passing after 2 iterations\n       → Code review complete - 0 issues\n       → PR ready: https://github.com/you/project/pull/47\n```\n\nLooking for the original Python-based Archon (task management + RAG)? It's fully preserved on the [ archive/v1-task-management-rag](https://github.com/coleam00/Archon/tree/archive/v1-task-management-rag) branch.\n\nMost users should start with the- it walks you through credentials, installs the Archon skill into your projects, and gives you the web dashboard.[Full Setup]\n\nAlready have Claude Code and just want the CLI?Jump to the[Quick Install].\n\nClone the repo and use the guided setup wizard. This configures credentials, platform integrations, and copies the Archon skill into your target projects.\n\n**Prerequisites** - Bun, Claude Code, and the GitHub CLI\n\n**Bun** - [bun.sh](https://bun.sh)\n\n```\n# macOS/Linux\ncurl -fsSL https://bun.sh/install | bash\n\n# Windows (PowerShell)\nirm bun.sh/install.ps1 | iex\n```\n\n**GitHub CLI** - [cli.github.com](https://cli.github.com/)\n\n```\n# macOS\nbrew install gh\n\n# Windows (via winget)\nwinget install GitHub.cli\n\n# Linux (Debian/Ubuntu)\nsudo apt install gh\n```\n\n**Claude Code** - [claude.ai/code](https://claude.ai/code)\n\n```\n# macOS/Linux/WSL\ncurl -fsSL https://claude.ai/install.sh | bash\n\n# Windows (PowerShell)\nirm https://claude.ai/install.ps1 | iex\ngit clone https://github.com/coleam00/Archon\ncd Archon\nbun install\nclaude\n```\n\nThen say: **\"Set up Archon\"**\n\nThe setup wizard walks you through everything: CLI installation, authentication, platform selection, and copies the Archon skill to your target repo.\n\nAlready have Claude Code set up? Install the standalone CLI binary and skip the wizard.\n\n**macOS / Linux**\n\n```\ncurl -fsSL https://archon.diy/install | bash\n```\n\nx64 compatibility:The macOS/Linux quick install requires AVX2 on x64 CPUs. Older Intel/AMD hardware and virtual machines that mask AVX2 should use the[source installation guide]. ARM64 quick installs are unaffected.\n\n**Windows (PowerShell)**\n\n```\nirm https://archon.diy/install.ps1 | iex\n```\n\n**Homebrew**\n\n```\nbrew install coleam00/archon/archon\n```\n\nCompiled binaries need aThe quick-install binaries don't bundle Claude Code. Install it separately, then point Archon at it:`CLAUDE_BIN_PATH`\n\n.\n\n```\n# macOS / Linux / WSL\ncurl -fsSL https://claude.ai/install.sh | bash\nexport CLAUDE_BIN_PATH=\"$HOME/.local/bin/claude\"\n\n# Windows (PowerShell)\nirm https://claude.ai/install.ps1 | iex\n$env:CLAUDE_BIN_PATH = \"$env:USERPROFILE\\.local\\bin\\claude.exe\"\n```\n\nOr set\n\n`assistants.claude.claudeBinaryPath`\n\nin`~/.archon/config.yaml`\n\n. The Docker image ships Claude Code pre-installed. See[AI Assistants → Binary path configuration]for details.\n\nOnce you've completed either setup path, go to your project and start working:\n\n```\ncd /path/to/your/project\nclaude\nUse archon to fix issue #42\nWhat archon workflows do I have? When would I use each one?\n```\n\nThe coding agent handles workflow selection, branch naming, and worktree isolation for you. Projects are registered automatically the first time they're used.\n\nImportant:Always run Claude Code from your target repo, not from the Archon repo. The setup wizard copies the Archon skill into your project so it works from there.\n\nArchon includes a web dashboard for chatting with your coding agent, running workflows, and monitoring activity. Binary installs: run `archon serve`\n\nto download and start the web UI in one step. From source: ask your coding agent to run the frontend from the Archon repo, or run `bun run dev`\n\nfrom the repo root yourself.\n\nRegister a project by clicking **+** next to \"Project\" in the chat sidebar - enter a GitHub URL or local path. Then start a conversation, invoke workflows, and watch progress in real time.\n\n**Key pages:**\n\n**Chat**- Conversation interface with real-time streaming and tool call visualization** Dashboard**- Mission Control for monitoring running workflows, with filterable history by project, status, and date** Workflow Builder**- Visual drag-and-drop editor for creating DAG workflows with loop nodes** Workflow Execution**- Step-by-step progress view for any running or completed workflow\n\n**Monitoring hub:** The sidebar shows conversations from **all platforms** - not just the web. Workflows kicked off from the CLI, messages from Slack or Telegram, GitHub issue interactions - everything appears in one place.\n\nSee the [Web UI Guide](https://archon.diy/adapters/web/) for full documentation.\n\nArchon ships with workflows for common development tasks:\n\n| Workflow | What it does |\n|---|---|\n`archon-assist` |\nGeneral Q&A, debugging, exploration - full Claude Code agent with all tools |\n`archon-fix-github-issue` |\nClassify issue → investigate/plan → implement → validate → PR → smart review → self-fix |\n`archon-create-issue` |\nClassify problem → gather context → investigate → create GitHub issue |\n`archon-issue-review-full` |\nComprehensive fix + full multi-agent review pipeline for GitHub issues |\n`archon-piv-loop` |\nGuided Plan-Implement-Validate loop with human review between iterations |\n`archon-idea-to-pr` |\nFeature idea → plan → implement → validate → PR → 5 parallel reviews → self-fix |\n`archon-plan-to-pr` |\nExecute existing plan → implement → validate → PR → review → self-fix |\n`archon-feature-development` |\nImplement feature from plan → validate → create PR |\n`archon-adversarial-dev` |\nBuild a complete application from scratch using adversarial development |\n`archon-smart-pr-review` |\nClassify PR complexity → run targeted review agents → synthesize findings |\n`archon-comprehensive-pr-review` |\nMulti-agent PR review (5 parallel reviewers) with automatic fixes |\n`archon-validate-pr` |\nThorough PR validation testing both main and feature branches |\n`archon-architect` |\nArchitectural sweep, complexity reduction, codebase health improvement |\n`archon-refactor-safely` |\nSafe refactoring with type-check hooks and behavior verification |\n`archon-interactive-prd` |\nCreate a PRD through guided conversation |\n`archon-ralph-dag` |\nPRD implementation loop - iterate through stories until done |\n`archon-workflow-builder` |\nGenerate a new Archon workflow YAML for your project |\n`archon-remotion-generate` |\nGenerate or modify Remotion video compositions with AI |\n`archon-resolve-conflicts` |\nDetect merge conflicts → analyze both sides → resolve → validate → commit |\n\nArchon ships 19 default workflows - run `archon workflow list`\n\nor describe what you want and the router picks the right one.\n\n**Or define your own.** Default workflows are great starting points - copy one from `.archon/workflows/defaults/`\n\nand customize it. Workflows are YAML files in `.archon/workflows/`\n\n, commands are markdown files in `.archon/commands/`\n\n. Same-named files in your repo override the bundled defaults. Commit them - your whole team runs the same process.\n\nSee [Authoring Workflows](https://archon.diy/guides/authoring-workflows/) and [Authoring Commands](https://archon.diy/guides/authoring-commands/).\n\nThe Web UI and CLI work out of the box. Optionally connect a chat platform for remote access:\n\n| Platform | Setup time | Guide |\n|---|---|---|\nTelegram |\n5 min |\n|\n\n**Slack**[Slack Guide](https://archon.diy/adapters/slack/)** GitHub Webhooks**[GitHub Guide](https://archon.diy/adapters/github/)** Discord**[Discord Guide](https://archon.diy/adapters/community/discord/)\n\n```\n┌─────────────────────────────────────────────────────────┐\n│  Platform Adapters (Web UI, CLI, Telegram, Slack,       │\n│                    Discord, GitHub)                     │\n└──────────────────────────┬──────────────────────────────┘\n                           │\n                           ▼\n┌─────────────────────────────────────────────────────────┐\n│                     Orchestrator                        │\n│          (Message Routing & Context Management)         │\n└─────────────┬───────────────────────────┬───────────────┘\n              │                           │\n      ┌───────┴────────┐          ┌───────┴────────┐\n      │                │          │                │\n      ▼                ▼          ▼                ▼\n┌───────────┐  ┌────────────┐  ┌──────────────────────────┐\n│  Command  │  │  Workflow  │  │    AI Assistant Clients  │\n│  Handler  │  │  Executor  │  │   (Claude / Codex / Pi)  │\n│  (Slash)  │  │  (YAML)    │  │                          │\n└───────────┘  └────────────┘  └──────────────────────────┘\n      │              │                      │\n      └──────────────┴──────────────────────┘\n                           │\n                           ▼\n┌─────────────────────────────────────────────────────────┐\n│          SQLite / PostgreSQL (14 core tables)           │\n│  Codebases • Conversations • Sessions • Workflow Runs   │\n│   Isolation Environments • Messages • Workflow Events   │\n│    Users • User Identities • Workflow Node Sessions     │\n│         Codebase Env Vars • User GitHub Tokens          │\n│           User Provider Keys • User AI Prefs            │\n│          (+ Better Auth tables, Postgres only)          │\n└─────────────────────────────────────────────────────────┘\n```\n\nFull documentation is available at ** archon.diy/docs**.\n\n| Topic | Description |\n|---|---|\n|\n\n[The Book of Archon](https://archon.diy/book/)[CLI Reference](https://archon.diy/reference/cli/)[Authoring Workflows](https://archon.diy/guides/authoring-workflows/)[Authoring Commands](https://archon.diy/guides/authoring-commands/)[Configuration](https://archon.diy/reference/configuration/)[AI Assistants](https://archon.diy/getting-started/ai-assistants/)[Deployment](https://archon.diy/deployment/)[Architecture](https://archon.diy/reference/architecture/)[Troubleshooting](https://archon.diy/reference/troubleshooting/)**For AI tools:** Point your LLM at [ /llms.txt](https://archon.diy/llms.txt) for an index of all documentation,\n\n[for the complete docs in a single file, or](https://archon.diy/llms-full.txt)\n\n`/llms-full.txt`\n\n[for a condensed version.](https://archon.diy/llms-small.txt)\n\n`/llms-small.txt`\n\nArchon sends a few anonymous events so maintainers can see which workflows get real usage, on what platforms, and whether runs succeed — and prioritize accordingly. **No PII, ever.** Events: `archon_started`\n\n(once per CLI invocation / server boot), `archon_active`\n\n(daily heartbeat while a server is running, so long-running installs stay counted), `chat_turn_handled`\n\n(each direct AI chat turn — platform, provider, model, duration, and usage totals; never message content), `workflow_invoked`\n\n(each workflow start), `workflow_completed`\n\n/ `workflow_failed`\n\n(each run outcome), `workflow_approval_resolved`\n\n(each human approve/reject decision — the binary resolution only, never comments or reasons), and `codebase_registered`\n\n(a pure count when a project is registered — no name, path, or URL).\n\n**What's collected (categorical only):**\n\n**Workflow name**— the real name for*bundled*(Archon-authored) workflows;`\"custom\"`\n\nfor your own workflows, so private names never leave your machine.**Run shape & outcome**— platform (`cli`\n\n/`web`\n\n/`slack`\n\n/…), provider id (plus the model id on`workflow_invoked`\n\n), node count, which node types and features are used (loop/approval/script/bash, structured output, persisted sessions, MCP, skills, fresh-context loops), success/failure, duration, a categorical failure reason, and a fixed-enum failure class (`fatal`\n\n/`transient`\n\n/`unknown`\n\n— never raw error text) plus the failed node's type.**Chat activity**— one event per direct-chat AI turn with platform, provider, model, duration, and completed/failed. Message content, prompts, and conversation ids are never sent.**Aggregate usage**— provider-reported token counts and cost (USD) per workflow run and chat turn, plus total loop iterations per run. Numeric totals only — never the content the tokens represent.**Machine context**— OS, architecture, Archon version, runtime, whether it's a binary build, and a CI flag.** Deployment shape**(server only) — which adapters are enabled (booleans), database kind (`sqlite`\n\n/`postgresql`\n\n), whether web auth and multi-user mode are on, and the GitHub auth mode. Configuration*values*(tokens, URLs, hosts) are never sent.- A random install UUID stored at\n`~/.archon/telemetry-id`\n\n. Nothing else.\n\n**What's not collected:** your code, prompts, messages, custom workflow names, workflow descriptions, git remotes, file paths, usernames, tokens, AI output, error message text, your IP address, your geographic location — none of it.\n\n**Opt out:** set any of these in your environment:\n\n```\nARCHON_TELEMETRY_DISABLED=1\nDO_NOT_TRACK=1        # de facto standard honored by Astro, Bun, Prisma, Nuxt, etc.\nPOSTHOG_API_KEY=off   # off | 0 | false | disabled | \"\" all disable\n```\n\nCI environments (`CI=true`\n\n) are auto-disabled — forks running fixtures in GitHub Actions, CircleCI, etc. do not send events.\n\n**Check the current state:** run `archon telemetry status`\n\nto see whether telemetry is enabled, why (if not), the install UUID, and the active host. Run `archon telemetry reset`\n\nto rotate the install UUID. `archon doctor`\n\nalso surfaces the current state in its check list.\n\nSelf-host PostHog or use a different project by setting `POSTHOG_API_KEY`\n\nand `POSTHOG_HOST`\n\n.\n\nContributions welcome! See the open [issues](https://github.com/coleam00/Archon/issues) for things to work on.\n\nPlease read [CONTRIBUTING.md](/coleam00/Archon/blob/dev/CONTRIBUTING.md) before submitting a pull request.", "url": "https://wpnews.pro/news/open-source-harness-builder-for-ai-coding", "canonical_source": "https://github.com/coleam00/Archon", "published_at": "2026-08-12 05:52:54+00:00", "updated_at": "2026-08-12 06:11:48.568424+00:00", "lang": "en", "topics": ["developer-tools", "ai-agents", "ai-tools"], "entities": ["Archon", "coleam00", "Claude Code", "GitHub", "Bun", "Slack", "Telegram"], "alternates": {"html": "https://wpnews.pro/news/open-source-harness-builder-for-ai-coding", "markdown": "https://wpnews.pro/news/open-source-harness-builder-for-ai-coding.md", "text": "https://wpnews.pro/news/open-source-harness-builder-for-ai-coding.txt", "jsonld": "https://wpnews.pro/news/open-source-harness-builder-for-ai-coding.jsonld"}}