wshobson/agents

**Repository**: https://github.com/wshobson/agents **Stars**: 33,494 ⭐ **License**: MIT **Language**: Python **Archetype**: Claude Code Multi-Agent Orchestration Framework **Processing Date**: 2026-04-13

Overview

A comprehensive production-ready system combining 182 specialized AI agents, 16 multi-agent workflow orchestrators, 149 agent skills, and 96 commands organized into 77 focused, single-purpose plugins for Claude Code. This represents one of the largest organized collections of agent orchestration patterns and workflows.

Repository Structure

wshobson/agents/
├── .claude-plugin/          # Claude Code plugin configuration
├── plugins/                 # 77 focused plugins (main content)
│   ├── accessibility-compliance/
│   ├── agent-orchestration/
│   └── ...
├── docs/                    # Comprehensive documentation
│   ├── plugins.md           # Plugin reference
│   ├── agents.md            # Agent reference
│   ├── agent-skills.md      # Skills documentation
│   ├── usage.md             # Usage patterns
│   └── architecture.md      # Design principles
└── tools/                   # Supporting tooling

Key Features & Architectural Patterns

Granular Plugin Architecture

• **77 focused plugins** optimized for minimal token usage and composability
• Average 3.6 components per plugin (follows Anthropic's 2-8 pattern)
• Each plugin completely isolated with its own agents, commands, and skills
• Install only what you need - no unnecessary resources loaded into context

Agent Orchestration Models

• **182 specialized agents** across domain expertise areas
• **16 workflow orchestrators** for complex multi-agent operations
• **Progressive disclosure** - skills load knowledge only when activated
• **Three-tier model strategy** for Opus 4.6, Sonnet 4.6 & Haiku 4.5

Skills System (149 Skills)

• Modular knowledge packages with progressive disclosure
• Token efficiency through selective loading
• Specialized expertise across domains
• Compatible with agent orchestration patterns

Notable Orchestration Patterns

Multi-Agent Review Systems

• comprehensive-review plugin with architect-review, code-reviewer, security-auditor
• Parallel analysis with different perspectives
• Coordinated feedback synthesis

Full-Stack Orchestration

• full-stack-orchestration plugin for complex development workflows
• Multi-agent coordination for end-to-end development
• Phase-based execution with handoffs

Agent Teams (Experimental)

• Parallel code review with specialized reviewers
• Hypothesis-driven debugging with multiple agents
• Research teams with parallel investigation streams

Context-Driven Development (Conductor Plugin)

• Structured Context → Spec & Plan → Implement workflow
• Track-based development with semantic organization
• TDD workflow with verification checkpoints

Quality Framework (PluginEval)

Three-layer evaluation framework:

• **Static analysis** (instant)
• **LLM judge** (semantic)
• **Monte Carlo simulation** (statistical)

Quality dimensions:

• Triggering accuracy, orchestration fitness, output quality
• Scope calibration, progressive disclosure, token efficiency
• Robustness, structural completeness, ecosystem coherence

Technology Integration

• **Claude Code** as primary harness
• **Agent Skills** for knowledge management
• **Plugin marketplace** distribution model
• **Makefile** for development automation
• **CI/CD** integration patterns

Extractable Processes

Core Orchestration Patterns

1. **Multi-perspective review processes** - parallel expert analysis 2. **Progressive skill disclosure** - token-efficient knowledge loading 3. **Plugin composition strategies** - modular system assembly 4. **Agent handoff protocols** - structured workflow transitions

Development Methodologies

1. **Context-driven development** - vision → spec → implementation 2. **Track-based organization** - semantic work unit management 3. **Verification checkpoint patterns** - quality gates in workflows 4. **Semantic revert strategies** - logical undo operations

Quality Assurance Frameworks

1. **Multi-layer evaluation** - static, semantic, statistical analysis 2. **Anti-pattern detection** - automated quality assessment 3. **Statistical rigor** - Wilson score CI, bootstrap validation 4. **CI gate integration** - threshold-based quality control

Value for Babysitter Marketplace

**High Value** - This repository contains the most comprehensive collection of agent orchestration patterns discovered, with well-documented architectural principles and proven workflows. The progressive disclosure patterns, multi-agent coordination strategies, and quality evaluation frameworks provide substantial value for babysitter process development.

Key Extractable Elements

• Multi-agent orchestration patterns for complex workflows
• Progressive disclosure mechanisms for token efficiency
• Quality evaluation frameworks with statistical rigor
• Plugin composition and modular architecture principles

Classification

**Category**: Multi-Agent Orchestration Framework **Subcategory**: Claude Code Plugin Ecosystem **Complexity**: Enterprise-grade **Maturity**: Production-ready **Adoption**: High (33K+ stars, active development)

Research Notes

This repository represents the current state-of-the-art in Claude Code agent orchestration, with sophisticated approaches to:

• Token efficiency through modular loading
• Multi-agent coordination patterns
• Quality measurement and validation
• Enterprise-scale deployment patterns

The architectural principles and orchestration patterns here could significantly inform babysitter's multi-agent workflow capabilities.

Library Mapping

Plugin Marketplace Mapping