Skip to content

Repository files navigation

Waypoint

Modular Claude Code plugins organized into workflows and technologies.

Waypoint is a collection of Claude agents, commands, and skills organized into two categories:

  • Workflows: Operational tooling that enhances how you work with Claude Code (git worktree management)
  • Technologies: Domain knowledge for infrastructure tools (Terraform, Ansible, Docker, Proxmox)

Each category is a plugin with its own .claude-plugin/plugin.json manifest.

Philosophy

  • Workflows vs Technologies: Clear separation between operational tooling and domain knowledge
  • Modular: Each plugin is independent and can be installed separately
  • Namespaced: Agents, commands, and skills are organized by plugin
  • Reusable: Clone once, use everywhere via the Claude Code plugin system

Installation

Install Waypoint as a Claude Code plugin:

# Clone the repository
git clone git@github.com:poindexter12/waypoint.git

# The plugin system will automatically discover and load the plugins
# Just ensure the repository is accessible to Claude Code

Waypoint uses the Claude Code plugin system. Each subdirectory (workflows/, technologies/, claire/) contains a .claude-plugin/plugin.json manifest that defines the available agents, commands, and skills.

Available Plugins

Workflows

Operational tooling that enhances development workflows.

Agents: consultant - Git worktree strategy and organization Commands: /working-tree:new, /working-tree:status, /working-tree:list, /working-tree:destroy, /working-tree:adopt Skills: working-tree - Worktree patterns and templates

Technologies

Domain knowledge for infrastructure tools.

Agents:

  • terraform - Terraform/IaC patterns and best practices
  • ansible - Ansible playbooks and automation
  • docker-compose - Docker and container orchestration
  • proxmox - Proxmox virtualization platform

Skills:

  • terraform - HCL patterns, state management, Proxmox provider
  • ansible - Playbooks, inventory, modules
  • docker - Dockerfile, Compose, networking
  • proxmox - VMs, LXC, storage, clustering

Claire (Meta-Tooling)

Tools for creating and optimizing Claude Code components.

Commands: /claire:fetch-docs Agents: coordinator, author-agent, author-command Skills: doc-validator

Directory Structure

waypoint/
├── .claude-plugin/plugin.json    # Root plugin index
├── workflows/                    # Operational tooling
│   ├── .claude-plugin/plugin.json
│   ├── agents/consultant.md
│   ├── commands/*.md
│   └── skills/working-tree/
├── technologies/                 # Domain knowledge
│   ├── .claude-plugin/plugin.json
│   ├── agents/*.md               # terraform, ansible, docker, proxmox
│   └── skills/*/                 # Per-technology skills
└── claire/                       # Meta-tooling
    ├── .claude-plugin/plugin.json
    ├── agents/
    ├── commands/
    └── skills/

Makefile Reference

The Makefile provides tools for managing plugin manifests and versions.

Targets

Target Description
help Show all available commands (default)
manifest Update plugin.json files from directory contents
version V=x.y.z Bump version in all plugin.json files
changelog-preview Show commits since last version tag
clean Remove build artifacts

Examples

# Update all plugin.json manifests from directory contents
make manifest

# Bump all versions to 1.2.0
make version V=1.2.0

# Preview unreleased changes for changelog
make changelog-preview

Development

Adding Components

Add a new workflow agent/command:

  • Add agent to workflows/agents/
  • Add commands to workflows/commands/
  • Add skill (if needed) to workflows/skills/

Add a new technology:

  • Add agent to technologies/agents/my-tech.md
  • Add skill to technologies/skills/my-tech/SKILL.md
  • Add references to technologies/skills/my-tech/references/

Then run make manifest to update plugin.json files.

Plugin Structure

Each plugin category follows this pattern:

plugin-category/
├── .claude-plugin/
│   └── plugin.json      # Plugin manifest (auto-updated by make manifest)
├── agents/
│   └── *.md            # Agent definitions
├── commands/
│   └── *.md            # Command definitions (workflows only)
└── skills/
    └── skill-name/     # Skill directories
        ├── SKILL.md    # Required: skill definition
        ├── REFERENCE.md # Optional: detailed documentation
        └── references/ # Optional: additional docs

Slash Command and Agent File Format

Slash Command Files (commands/*.md) use YAML frontmatter with these allowed fields:

  • description: Brief description of the command (defaults to first line of prompt)
  • argument-hint: Arguments expected, shown during auto-completion (e.g., <branch> [--mode <mode>])
  • allowed-tools: List of tools the command can use (inherits from conversation if not specified)
  • model: Specific model to use (inherits from conversation if not specified)
  • disable-model-invocation: Prevent SlashCommand tool from calling this command
  • Additional custom fields as needed (e.g., agent for delegation)

Agent Files (agents/*.md) use YAML frontmatter with these fields:

  • name (required): Unique identifier using lowercase letters and hyphens
  • description (required): Natural language description of the agent's purpose
  • tools (optional): Comma-separated list of specific tools (inherits all if omitted)
  • model (optional): Model alias (sonnet, opus, haiku) or inherit (defaults to configured subagent model)
  • permissionMode (optional): Permission handling (default, acceptEdits, bypassPermissions, plan, ignore)
  • skills (optional): Comma-separated list of skill names to auto-load

Skill Files (skills/skill-name/SKILL.md) use YAML frontmatter with these fields:

  • name (required): Unique identifier using lowercase letters and hyphens (max 64 chars)
  • description (required): Natural language description including trigger keywords (max 1024 chars)
  • allowed-tools (optional): Comma-separated list of tools the skill can use

Example slash command:

---
description: Create a new git worktree with AI context
argument-hint: <branch> [--mode <mode>] [--description "<text>"]
agent: working-tree-manager
---

Command instructions here...

Example agent:

---
name: working-tree-manager
description: Manages git worktrees with AI context tracking
tools: Bash, Read, Write, Grep
model: sonnet
---

You are a git worktree manager. When invoked...

Example skill:

---
name: doc-validator
description: Validate documentation files for completeness, accuracy, and consistency. Use when checking README files, API docs, or documentation quality.
allowed-tools: Read, Grep, Glob
---

# Documentation Validator Skill

Validates documentation files to ensure they are complete...

Contributing

Contributions are welcome! To contribute:

  1. Fork the repository
  2. Create a feature branch
  3. Add or improve a module
  4. Run make manifest to update plugin.json files
  5. Submit a pull request

License

MIT License - See LICENSE for details.

Credits

Created by Joe Seymour

Designed for use with Claude Code and compatible AI coding tools.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

Packages

Used by

Contributors

Languages