autopilot-cli 1.1.2

Autonomous multi-agent development orchestrator — task management, anti-pattern enforcement, hive-mind coordination, and interactive REPL from the terminal.

Autopilot CLI

Autonomous multi-agent development orchestrator. Autopilot CLI manages multi-agent development workflows with task management, anti-pattern enforcement, hive-mind coordination, deployment monitoring, and interactive REPL — all from the terminal.

Features

• Task Management — Sprint planning, task tracking, and velocity reporting with Fibonacci point estimation
• Anti-Pattern Enforcement — 11-category detection engine across 5 enforcement layers (editor, pre-commit, CI, guardrails, protected regions)
• Hive-Mind Orchestration — Spawn multi-agent sessions via claude-flow with Jinja2 objective templates, resource gating, and session tracking
• Agent Coordination — Document-mediated communication via markdown board files, question queues, and decision tracking
• Session Lifecycle — Daemon-based autonomous sessions with usage quotas, circuit breakers, and workspace isolation
• Deployment Monitoring — Health checking for services with Render integration
• Interactive REPL — Rich terminal UI with prompt-toolkit and command history
• Configuration — Pydantic v2 models with three-level YAML merge (global, project, CLI)

Installation

pip install autopilot-cli

Or with uv:

uv pip install autopilot-cli

Requires Python 3.12+.

Quick Start

# Initialize a project
autopilot init --name my-project

# Start an autonomous session
autopilot start

# Run a single cycle
autopilot cycle

# Enter interactive REPL
autopilot

# Show all commands
autopilot --help

CLI Commands

Command	Description
autopilot init	Initialize a new project
autopilot start	Start an autonomous session daemon
autopilot stop	Stop the running session
autopilot cycle	Run a single scheduler cycle
autopilot watch	Live dashboard
autopilot task	Task management and sprint planning
autopilot session	Session lifecycle (start, stop, pause, resume, list)
autopilot enforce	Anti-pattern enforcement engine
autopilot hive	Hive-mind orchestration (spawn, status, stop)
autopilot project	Project registry management
autopilot config	Configuration management
autopilot report	Reporting and analytics

Development

This project uses just as a task runner and uv for package management.

Requirements

• Python 3.12+
• uv
• just

Setup

git clone https://github.com/mandelbro/autopilot-cli.git
cd autopilot-cli
asdf install # installs python, uv, and just
just init

Running locally

# Run the CLI via just (pass args after --)
just run -- --help
just run -- init --name my-project
just run -- hive spawn tasks/tasks-1.md --task-ids 001-008 --dry-run

# Or directly with uv
uv run autopilot --help

Commands

# Run all checks (format, lint, typecheck, test)
just

# Individual commands
just test          # uv run pytest
just lint          # uv run ruff check --fix src/ tests/
just format        # uv run ruff format src/ tests/
just typecheck     # uv run pyright
just coverage      # pytest --cov --cov-fail-under=80

Project Structure

src/autopilot/
  cli/           # Typer CLI application and display helpers
  core/          # Configuration (Pydantic), data models, project management
  coordination/  # Board management, question queue, announcements
  enforcement/   # Anti-pattern enforcement engine and rules
  monitoring/    # Deployment health checking
  orchestration/ # Agent dispatch, scheduling, hive-mind coordination
  reporting/     # Metrics and reports
  uat/           # User acceptance testing pipeline
  utils/         # Shared utilities (subprocess, paths, git, sanitizer)

License

Apache 2.0 — see LICENSE for details.