rr

name	rr
description	Sync code and run commands on remote machines. Use when running tests, builds, or commands remotely, syncing files to hosts, setting up remote development, or troubleshooting rr configuration.
user-invocable	true
allowed-tools	["Bash","Read","Edit","Grep","Glob"]

rr (Road Runner) CLI

rr syncs code to remote machines and runs commands there. Handles host failover, file sync with rsync, distributed locking, and test output formatting.

Quick Reference

rr run "make test"     # Sync files + run command
rr exec "git status"   # Run command without syncing
rr sync                # Just sync files
rr <taskname>          # Run named task from config
rr provision           # Install missing tools on hosts
rr doctor              # Diagnose issues
rr monitor             # TUI dashboard for host metrics

Two-Config System

rr uses two config files:

Config	Location	Purpose
Global	~/.rr/config.yaml	Personal host definitions (SSH, directories)
Project	.rr.yaml	Shareable project settings (tasks, sync rules)

See config.md for complete config reference.

Minimal Global Config

version: 1
hosts:
  mini:
    ssh: [mac-mini.local, mac-mini-tailscale]
    dir: ${HOME}/projects/${PROJECT}

Minimal Project Config

version: 1
hosts: [mini]

sync:
  exclude: [.git/, node_modules/, .venv/]

tasks:
  test:
    run: pytest -v

Commands Overview

Command	Purpose
rr run "cmd"	Sync files, then run command
rr exec "cmd"	Run command without syncing
rr sync	Just sync files
rr <taskname>	Run named task
rr tasks	List available tasks
rr provision	Install missing tools on hosts
rr doctor	Diagnose issues
rr host list/add/remove	Manage hosts

See commands.md for full command reference.

Common Flags

• --pretty / -p - Opt into human-readable output (spinners, colors). Default is structured JSON.
• --host <name> - Target specific host
• --tag <tag> - Select host by tag
• --local - Force local execution
• --skip-requirements - Skip requirement checks

Run rr --help or rr <command> --help for complete flag reference.

Tasks

Define reusable commands in .rr.yaml:

tasks:
  test:
    description: Run tests
    run: pytest -v

  deploy:
    steps:
      - name: Build
        run: make build
      - name: Deploy
        run: ./deploy.sh

Run with: rr test, rr deploy

Extra arguments append to single-command tasks: rr test -k "test_login"

See tasks.md for parallel tasks, multi-step tasks, and advanced configuration.

Task Types and Arguments

Tasks come in three types. The type determines whether extra args work:

Type	Config field	Accepts args?	Example
Single-command	run:	YES	rr test -k "test_foo"
Multi-step	steps:	NO (errors)	rr deploy
Parallel	parallel:	NO unless forward_args: true	rr test-all

When you need custom args on a parallel task, bypass it:

# Preferred: use --cwd for subdirectory execution (path-traversal safe)
rr run --cwd backend "uv run pytest tests/bond/ -v"

# Fallback: manual cd pattern (avoid — quoting errors are common)
rr run "cd backend && uv run pytest tests/bond/ -v"

To check which type a task is: rr <task> --help

To forward args to all subtasks in a parallel task, set forward_args: true in the task config:

tasks:
  test-backend:
    parallel: [test-backend-api, test-backend-services]
    forward_args: true   # rr test-backend -k bond works

Choosing the Right Command

Need to run something remotely?
├── Named task exists? → rr <task>
│   ├── Single-command task → rr <task> <args>   (args forwarded)
│   ├── Parallel task (forward_args: true) → rr <task> <args>
│   └── Parallel task (default) → rr run "cd <dir> && <cmd> <args>"
├── No task, files may have changed → rr run "<command>"   (syncs first)
└── Files already synced → rr exec "<command>"   (faster, skips sync)

Never nest rr inside rr exec — rr may not be installed on the remote:

# Wrong:  rr exec "rr sync && pytest"
# Right:  rr run "pytest"

Reading rr Output

rr sends phase events (connect, sync, exec) to stderr and command output to stdout.

rr test-opendata 2>/dev/null         # suppress all phase events
rr test-opendata --no-phases         # suppress intermediate events, keep final result JSON
rr test-opendata 2>&1                # capture everything (mixes phase JSON into output stream)

The final result is always a JSON line on stderr with "type":"result" containing exit code, host, and duration.

Remote Environment Bootstrap

Declare required tools with require: - rr verifies they exist before running commands:

# .rr.yaml
require:
  - go
  - node

tasks:
  build:
    run: make build
    require: [cargo]  # Task-specific requirement

# ~/.rr/config.yaml
hosts:
  gpu-box:
    ssh: [gpu.local]
    require: [nvidia-smi, python3]  # Host-specific requirements

Run with: rr test-all, rr quick-check

Setup Phase (Once Per Host)

Avoid redundant setup work (dependency sync, migrations) when multiple subtasks run on the same host:

tasks:
  test-all:
    setup: pip install -r requirements.txt   # Runs once per host
    parallel:
      - test-unit
      - test-integration
      - test-e2e

Setup runs exactly once per host before any subtasks execute. If a host runs 3 subtasks, setup runs once (not 3 times).

Parallel Task Flags

Flag	Purpose
--stream	Show real-time interleaved output with [host:task] prefixes
--verbose	Show full output per task on completion
--quiet	Summary only
--fail-fast	Stop on first failure (overrides config)
--max-parallel N	Limit concurrent tasks
--dry-run	Show plan without executing
--local	Force local execution (no remote hosts)

Task Dependencies

Define task execution order with depends. Tasks run their dependencies first, then execute their own command:

tasks:
  lint:
    run: golangci-lint run
  test:
    run: go test ./...
  build:
    run: go build ./...

  # Linear chain: lint -> test -> build
  ci:
    description: Full CI pipeline
    depends:
      - lint
      - test
      - build

Run with: rr ci

Parallel Groups in Dependencies

Run multiple dependencies simultaneously:

tasks:
  lint:
    run: golangci-lint run
  typecheck:
    run: mypy .
  test:
    run: pytest

  ci:
    depends:
      - parallel: [lint, typecheck]  # Run simultaneously
      - test                          # Run after parallel completes

Executes: [lint, typecheck] (parallel) -> test

Orchestrator Tasks

Tasks with only depends orchestrate without running their own command:

tasks:
  lint:
    run: golangci-lint run
  test:
    run: go test ./...

  verify:
    description: Run all checks
    depends: [lint, test]
    # No 'run' - just orchestrates

Dependency Flags

Flag	Purpose
--skip-deps	Skip dependencies, run only the target task
--from <task>	Start from a specific task in the chain

rr ci                  # Full dependency chain
rr ci --skip-deps      # Only run ci task itself
rr ci --from test      # Start from test, skip lint

Dependency Features

• Deduplication: Tasks run once even if referenced multiple times (diamond deps)
• Validation: Circular dependencies detected at config load
• Fail-fast: Stops on first failure when fail_fast: true
• Timeout: Honor timeout field for entire dependency chain

Output Modes

• progress (default): Live status indicators with spinners
• stream: Real-time output with [host:task] prefixes
• verbose: Full output shown when each task completes
• quiet: Summary only at the end

Example:

rr test-all --stream    # See all output in real-time
rr test-all --dry-run   # Preview what would run
rr test-all --local     # Run locally without remote hosts

Missing tools trigger actionable error messages. Tools with built-in installers (40+) can be auto-installed.

See requirements.md for complete requirements reference.

How It Works

1. Host Selection: Tries SSH aliases in order until one connects
2. Requirements: Verifies required tools exist (if configured)
3. File Sync: Uses rsync with exclude/preserve patterns
4. Locking: Creates lock on remote; if locked, tries next host
5. Execution: Runs command with configured environment

Troubleshooting

Problem	Fix
SSH fails	Check ssh <alias> manually, verify ~/.ssh/config
"handshake failed" but ssh works	Key not in agent: ssh-add ~/.ssh/id_rsa, add AddKeysToAgent yes to SSH config
"command not found"	Add setup_commands or check require config
Sync slow	Add large dirs to sync.exclude
Lock stuck	rr unlock
Lock timeout after crash or cancel	rr unlock --all — stale locks from sessions that didn't exit cleanly

Self-Documentation

When you're unsure about a task's type, subtasks, or flags, run rr <task> --help before executing it. The help output shows whether the task is parallel (and lists subtasks), what flags it accepts, and whether it takes extra arguments. This is faster than reading .rr.yaml and avoids the silent-arg-drop problem.

See troubleshooting.md for detailed diagnostics.

Structured Output (Default)

rr defaults to structured output (agent-first). No flags needed. Phase events are emitted as JSON lines to stderr, command stdout/stderr passes through undecorated.

# Default behavior - structured JSON events on stderr, raw output on stdout
rr run "make test"
rr test

# Opt into human-readable spinners/colors
rr run --pretty "make test"

The --machine / -m flag still works but is a no-op (structured is already the default).

See machine-interface.md for JSON event format and error codes.

Quick Setup

# 1. Add a host
rr host add

# 2. Initialize project
cd your-project && rr init

# 3. Verify
rr doctor

# 4. Run
rr run "make test"

When to Use Each Command

Situation	Command
Run tests with latest code	rr run "make test"
Quick check on remote	rr exec "git log -1"
Prep remote before multiple runs	rr sync
Install missing tools on hosts	rr provision
Debug connection issues	rr doctor
Watch resource usage	rr monitor
First time setup	rr init
Add new machine	rr host add

Reference Files

• config.md - Complete config reference (global + project)
• commands.md - All commands and flags
• tasks.md - Task definitions, parallel execution, multi-step
• requirements.md - Remote environment bootstrap
• machine-interface.md - JSON output and error codes
• troubleshooting.md - Diagnostics and common fixes