Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/coleam00/Archon/llms.txt

Use this file to discover all available pages before exploring further.

Archon is an open-source workflow engine for AI coding agents. Where a plain AI session is unpredictable — the model might skip planning, forget to run tests, or produce a different structure each time — Archon lets you encode your entire development process as a YAML workflow that runs the same way every time. Think of it as what Dockerfiles did for infrastructure, or what GitHub Actions did for CI/CD, applied to AI-assisted software development.

​
The problem Archon solves

When you ask an AI agent to “fix this bug”, the outcome depends on the model’s mood. It might skip investigation. It might not run your test suite. It might write a PR description that ignores your template. Every run is different, and none of it is owned by you. Archon fixes this by separating structure from intelligence. You define the phases, validation gates, and handoffs in YAML. The AI fills in the intelligence at each node — planning, coding, reviewing — but the sequence is deterministic and version-controlled alongside your project.
.archon/workflows/build-feature.yaml
nodes:
  - id: plan
    prompt: "Explore the codebase and create an implementation plan"

  - id: implement
    depends_on: [plan]
    loop:
      prompt: "Read the plan. Implement the next task. Run validation."
      until: ALL_TASKS_COMPLETE
      fresh_context: true

  - id: run-tests
    depends_on: [implement]
    bash: "bun run validate"

  - id: review
    depends_on: [run-tests]
    prompt: "Review all changes against the plan. Fix any issues."

  - id: approve
    depends_on: [review]
    loop:
      prompt: "Present the changes for review. Address any feedback."
      until: APPROVED
      interactive: true

  - id: create-pr
    depends_on: [approve]
    prompt: "Push changes and create a pull request"

​
Key features

DAG workflows

Define multi-step coding tasks as directed acyclic graphs. Independent nodes run in parallel; dependent nodes wait. Fan out for parallel reviews, converge for final actions.

Git worktree isolation

Every workflow run gets its own git worktree. Run five parallel fixes without conflicts. Failed runs don’t leave your working branch dirty.

Multi-platform

Trigger workflows from the CLI, the web dashboard, Telegram, Slack, Discord, or GitHub webhooks. All conversations appear in one unified dashboard.

Multi-provider

Mix Claude Code, Codex, and Pi (20+ LLM backends) within a single workflow. Set a default per project or override per node.

​
How it works

Archon sits between you and your AI coding agent. When you run a workflow, Archon:
  1. Creates an isolated git worktree for the task
  2. Executes each DAG node in topological order (parallel where possible)
  3. At each AI node, invokes the configured provider with the resolved prompt
  4. Captures outputs and passes them to downstream nodes via $nodeId.output
  5. Runs bash or script nodes deterministically (no AI involved)
  6. Surfaces results, artifacts, and PR links back to you
Platform Adapters (Web UI, CLI, Telegram, Slack, Discord, GitHub)
                            │
                     Orchestrator
               (Routing & Context Management)
                 │                       │
         Command Handler          Workflow Executor
          (Slash cmds)              (YAML DAG)
                 │                       │
                 └───────────────────────┘
                            │
              SQLite / PostgreSQL (7 Tables)

​
Bundled workflows

Archon ships 17 production-ready workflows. A few highlights:
WorkflowWhat it does
archon-assistGeneral Q&A, debugging, exploration — full agent with all tools
archon-fix-github-issueClassify → investigate → implement → validate → PR → review → self-fix
archon-idea-to-prFeature idea → plan → implement → validate → PR → 5 parallel reviews → self-fix
archon-smart-pr-reviewComplexity-adaptive PR review, routes to relevant agents only
archon-resolve-conflictsDetect merge conflicts → analyze → resolve → validate → commit
You don’t need to remember workflow names. Describe what you want — “fix issue #42”, “review this PR” — and the router picks the best match. Run archon workflow list to see everything available.

​
Where to go next

Quickstart

Install Archon and run your first workflow in under five minutes.

Core concepts

Understand workflows, DAG nodes, isolation, providers, and adapters.

AI assistants

Set up Claude Code, Codex, or Pi as your AI backend.

Authoring workflows

Write custom YAML workflows tailored to your project.

Build docs developers (and LLMs) love

Get started for freeTalk to us