Skip to content

Home

Jump to bottom
Griffen Fargo edited this page Apr 30, 2026 · 23 revisions

🚀 Quick Start

New to coco? Start here for a complete walkthrough from installation to your first AI-generated commit.

📖 Getting Started Guide

  • Installation methods (npm, npx, global vs local)
  • First-time setup with AI provider selection
  • Your first commit walkthrough
  • Essential commands and workflows
  • Quick troubleshooting for beginners

⚙️ Configuration & Setup

Core Configuration

🔧 Configuration Overview

  • Complete configuration reference with all options
  • Multiple configuration methods (CLI, env vars, config files)
  • AI provider setup (OpenAI, Anthropic, Ollama)
  • Priority system and best practices

📁 Ignoring Files & Extensions

  • Advanced pattern matching with minimatch
  • Project-specific ignore configurations
  • Performance optimization through file filtering
  • Real-world examples for different project types

AI Providers

🏠 Using Ollama

  • Complete local AI setup guide
  • Model recommendations by hardware
  • Performance optimization and GPU acceleration
  • Privacy-focused development workflows

👥 Team & Enterprise

🤝 Team Collaboration

  • Shared configuration strategies
  • Git hooks and CI/CD integration
  • Security best practices for API keys
  • Team onboarding and training materials
  • Enterprise-scale deployment

🔧 Advanced Features

⚡ Advanced Usage

  • Custom prompts and templates
  • Multi-project configurations
  • Automation and scripting
  • Performance optimization techniques
  • Integration with development tools

🖥️ Coco UI

  • Use coco ui as a GitKraken-style terminal Git workstation
  • Browse history, inspect status, open dedicated diffs, and stage files/hunks
  • Compose commits manually or request explicit AI drafts after confirmation
  • Keep coco log -i as a supported alias into history mode

📜 Interactive Log TUI

  • History-focused guide for coco log -i and coco ui --view history
  • Search with fuzzy-ranked filtering and common lazygit/fzf keybindings
  • Inspect selected commit metadata, changed files, stats, and hunk previews

🧩 Commit Split

  • Plan smaller related commits from broad staged changes
  • Apply file-level and hunk-level split commit groups
  • Understand safety checks and fail-closed behavior

🧠 Dynamic Model Routing

  • Use service.model: "dynamic" with provider-specific model profiles
  • Route summarize, commit, changelog, review, recap, repair, and large-diff tasks
  • Customize model choices per task

📊 AI Call Audit

  • Inventory remote AI call paths
  • Understand task metadata, budget handling, and telemetry fields
  • Track model strategy follow-ups

🛡️ Command Reliability Audit

  • Review command coverage, CI gates, prompt budgets, and v1 reliability gaps
  • Track recommendations for future command work

📝 Documentation Workflow

  • Maintain the wiki locally from .wiki/
  • Keep the wiki as the canonical documentation source
  • Use repository docs only as temporary planning or migration material

🗺️ Product Roadmap

🧯 AI-Assisted Merge Conflict Resolution

  • Detect conflict state and explain current/incoming intent
  • Propose reviewable resolutions without blind writes
  • Build toward a conflict-focused interactive UI

🧹 AI Commit Range Recomposition

  • Analyze messy local commit ranges
  • Propose cleaner reviewable commit sequences
  • Keep safe, non-mutating plan mode as the first milestone

🧱 Worktree And Parallel Agent Workspaces

  • Show and manage local worktrees from Coco
  • Support agent-friendly workspace status output
  • Prepare for parallel branch, issue, and PR workflows

🆘 Help & Support

🔍 Troubleshooting Guide

  • Common issues and solutions
  • Diagnostic commands and debugging
  • Platform-specific problems (Windows, macOS, Linux)
  • API and connectivity issues
  • Performance troubleshooting

🎯 Use Cases

Individual Developers

  • Consistent Commits: Never write generic commit messages again
  • Learning Tool: Understand conventional commits through examples
  • Time Saving: Generate contextual messages in seconds
  • Offline Capable: Use Ollama for complete privacy and offline work

Development Teams

  • Standardization: Enforce consistent commit message formats
  • Onboarding: Help new team members learn project conventions
  • Code Review: Better commit messages improve review efficiency
  • Release Management: Conventional commits enable automated changelogs

Enterprise Organizations

  • Compliance: Audit trails and consistent documentation
  • Security: Local AI models keep code completely private
  • Cost Control: Shared Ollama servers eliminate per-request costs
  • Integration: Seamless integration with existing development workflows

🔗 Quick Links

Essential Commands

# Interactive commit (recommended)
coco -i

# Conventional commits format
coco --conventional

# Add context for complex changes
coco -a "Resolves login timeout issue"

# Include ticket from branch name
coco --append-ticket

# Generate changelog
coco changelog

# Open the Git workstation
coco ui

# Browse history interactively through the legacy entrypoint
coco log -i

# Summarize recent work
coco recap --yesterday

Configuration Examples

# Quick setup
coco init

# Project-specific setup
coco init --scope project

# Global user setup
coco init --scope global

Common Workflows

# Daily development
git add .
coco -i

# Team workflow with validation
git add .
coco --conventional -i

# Release workflow
coco changelog --since-last-tag

📖 Additional Resources

Community & Support

Development

🎉 Get Started Today

Ready to transform your git workflow? Choose your path:

This documentation is continuously updated. For the latest information, visit the GitHub repository or join our Discord community.

Clone this wiki locally