Skip to main content

Babysitter User Guide

Welcome to the Babysitter documentation. Babysitter is an AI-powered development workflow orchestrator that enables iterative quality convergence with human-in-the-loop approval through breakpoints.


Quick Start

Get up and running with Babysitter in minutes.

StepDescriptionTime
InstallationInstall the CLI and Claude Code plugin5 min
QuickstartConfigure your environment5 min
First RunExecute your first babysitter workflow10 min

What is Babysitter? (Start Here if You're New)

Babysitter automates the "try, check, fix, repeat" cycle of development so you don't have to do it manually.

The Problem Babysitter Solves

When you ask an AI to write code, you typically:

  1. Get code from the AI
  2. Run tests → tests fail
  3. Send errors back to AI
  4. Get fixed code
  5. Run tests again → still failing
  6. Repeat 5-10 times...

Babysitter automates this entire loop, running it until your quality targets are met.

How It Works (In Plain English)

┌─────────────────────────────────────────────────────────────────┐
│ YOU: "Build a login page with tests" │
│ ↓ │
│ BABYSITTER: Runs this loop automatically: │
│ 1. AI writes code │
│ 2. Tests run → 60% pass │
│ 3. AI fixes failures │
│ 4. Tests run → 85% pass │
│ 5. AI fixes remaining issues │
│ 6. Tests run → 95% pass ✓ Target met! │
│ ↓ │
│ YOU: Review and approve the final result │
└─────────────────────────────────────────────────────────────────┘

Key Terms You'll See

TermWhat It MeansExample
ProcessA workflow definition"Build feature with TDD"
RunOne execution of a processRunning the TDD workflow for your login page
TaskA single step in the process"Write tests", "Run linter", "Check coverage"
Quality GateA check that must passTests must be 90% passing
BreakpointA pause for human approval"Review this code before I deploy it" (handled in chat or via web UI)
IterationOne try-check-fix cycleAttempt #3 to pass the tests
ConvergenceImproving until target metGoing from 60% → 85% → 95%

Your First 5 Minutes

What you'll do:

  1. Install Babysitter (1 command)
  2. Run a simple workflow (1 command)
  3. See it iterate until tests pass
  4. Approve the result

What you'll learn:

  • How the iteration loop works
  • What a quality gate looks like
  • How to approve at breakpoints

What you'll see:

/babysitter:call build a calculator with add, subtract, multiply, divide using TDD

Creating run: calculator-20260125-143012
Process: TDD Quality Convergence
Target: 90% quality

Iteration 1: Quality 65/100 - Tests: 6/10 passing
→ AI fixing test failures...

Iteration 2: Quality 82/100 - Tests: 9/10 passing
→ AI improving code coverage...

Iteration 3: Quality 95/100 - Target met! ✅

Claude: The implementation is complete. Quality score: 95/100.
Do you approve the final result?
[Approve] [Request Changes]

You: [Approve]

Done! Your calculator module is ready.

Note: Breakpoints (approval prompts) are handled directly in the chat when using Claude Code. No external service needed!

The main command: /babysitter:call <your request> handles everything automatically.

Start the Quick Start Tutorial


Documentation Sections

Tutorials

Step-by-step learning guides that take you from beginner to expert.

TutorialLevelTimeDescription
Getting StartedBeginner20 minInstallation, setup, and your first run
Build a REST APIBeginner45 minCreate a complete REST API with TDD
Custom ProcessIntermediate60 minBuild your own process definition
Multi-Phase WorkflowsAdvanced90 minOrchestrate complex multi-phase development

Features

Deep dives into Babysitter's core capabilities.

FeatureDescription
Process Library2,239 JavaScript process files in the live generated snapshot, plus methodology, shared-process, skill, and agent layers discovered under library/
Two-Loops ArchitectureHybrid agentic systems - symbolic orchestration + agentic harness, guardrails, and evidence-driven completion
Quality ConvergenceFive quality gate types (tests, code quality, static analysis, security, performance) with 90-score patterns
Best PracticesFour guardrail layers, multi-gate validation, workflow design, and team collaboration patterns
BreakpointsHuman-in-the-loop approval system for critical decisions
Process DefinitionsCustomizable workflow templates and task orchestration
Journal SystemEvent-sourced audit trail and state reconstruction
Run ResumptionContinue interrupted workflows from any point
Parallel ExecutionConcurrent task execution for faster results

Highlight: The Process Library snapshot currently tracks 2,239 process files across 38 methodology families and the full specialization tree. Explore the library →

Essential Reading: Understanding the Two-Loops Architecture is key to designing reliable, bounded agentic workflows with proper guardrails and evidence-driven completion.


Reference

Technical specifications and lookup resources.

ReferenceDescription
Slash CommandsCore modes (call, yolo, forever, plan) and utility commands for Claude Code
CLI ReferenceComplete command-line interface documentation
Package & Plugin MapCanonical public/internal docs map for active packages, apps, and harness plugins
ConfigurationEnvironment variables and config file options
Error CatalogAll error codes with solutions
GlossaryTerminology and definitions
FAQFrequently asked questions
TroubleshootingCommon issues and resolutions

Learning Paths

Choose a path based on your role and goals.

For Developers New to Babysitter

Start here if this is your first time using Babysitter:

  1. First: Read the "What is Babysitter?" section above - it takes 2 minutes and explains the core concepts
  2. Then: Complete the Getting Started tutorial (20 min) - you'll install and run your first workflow
  3. Practice: Build your first project with REST API Tutorial (45 min)
  4. Reference: Use the Glossary when you encounter unfamiliar terms (it has a quick-reference table at the top)

For Experienced Developers

  1. Quick setup via Installation
  2. Learn the Five Quality Gate Types for robust validation
  3. Study Best Practices for workflow design
  4. Reference the CLI for automation

For Technical Leads and Architects

  1. Start here: Understand the Two-Loops Architecture philosophy
  2. Study Quality Convergence for the 90-score convergence pattern
  3. Review the Four Guardrail Layers for safety and control
  4. Learn Journal System for audit compliance
  5. Explore Custom Process for team workflows

For Quality Engineers

  1. Essential: Study the Five Quality Gate Types
  2. Review The 90-Score Convergence Pattern
  3. Understand Evidence-Driven Completion
  4. Apply Domain-Specific Targets from Best Practices

For DevOps and Automation Engineers

  1. Install using Quickstart
  2. Master the CLI Reference
  3. Configure via Configuration Reference
  4. Automate with Run Resumption

What's New

Version 5.0.0

  • Unified the public npm surface around @a5c-ai/babysitter for the main CLI
  • Split optional runtime orchestration into @a5c-ai/babysitter-agent
  • Refreshed user-facing docs to match the current package and command boundaries

Recent Updates

VersionDateHighlights
5.0.02026-04-25CLI/runtime package split clarified across public docs

For the complete changelog, see the GitHub Releases.


Search Tips

Finding what you need quickly:

  • Commands: Search for the command name (e.g., run:create, effects:get)
  • Errors: Search for the error code or key words from the message
  • Concepts: Use terms from the Glossary
  • Tasks: Search for what you want to do (e.g., "resume", "breakpoint", "quality")

Getting Help

Documentation Resources

Community and Support


Documentation Structure

This documentation follows the Diataxis framework:

CategoryPurposeUser Mode
TutorialsLearning through guided projectsStudy
FeaturesUnderstanding capabilitiesStudy
ReferenceTechnical lookup informationWork
How-to GuidesTask-focused problem solvingWork

Contributing

Found an issue with the documentation? Contributions are welcome.

  1. Check existing issues first
  2. Submit corrections via pull request
  3. Follow the documentation style guide

Last updated: 2026-01-25