Agent Patterns
- Updated April 1, 2026
- ats-kinoshita-iso/agent-workshop
agent-patterns bundles skills for designing and evaluating multi-agent systems, including orchestrator/worker decomposition, output quality review, and self-improving evaluator/optimizer loops. It gives developers proven patterns for structuring and improving agent architectures.
Key points
- Orchestrator/worker decomposition
- Output quality review
- Evaluator/optimizer loops
- Multi-agent design
Agent Patterns by the numbers
- Data as of Jul 7, 2026 (Skillselion catalog sync)
/plugin marketplace add ats-kinoshita-iso/agent-workshop/plugin install agent-patterns@agent-workshopAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Last updated | April 1, 2026 |
|---|---|
| Repository | ats-kinoshita-iso/agent-workshop ↗ |
What it does
Design and evaluate multi-agent systems: orchestrator/worker decomposition, output quality review, and self-improving evaluator/optimizer loops.
README.md
agent-patterns
Skills for designing and evaluating multi-agent systems, derived from the anthropic-cookbook agent patterns.
Skills
/agent-plan
Decompose a task into an orchestrator + worker architecture. The orchestrator manages state and delegates subtasks to narrowly-scoped workers. Produces a concrete decomposition diagram and interface contracts between components.
/agent-review
Evaluate the output quality of an agent or pipeline run. Scores output on correctness, format compliance, completeness, and safety. Produces a structured review with specific improvement suggestions.
/agent-loop
Design a self-improving evaluator/optimizer loop. The loop generates candidate outputs, evaluates them against defined criteria, and iteratively refines the generation prompt until quality thresholds are met.
Installation
Add this plugin via the Claude Code plugin marketplace or copy the skills/ directory
into your project.
Sources
Derived from:
patterns/agents/orchestrator_workers.ipynb(anthropic-cookbook)patterns/agents/evaluator_optimizer.ipynb(anthropic-cookbook)patterns/agents/agent_loop.py(anthropic-cookbook)
Related skills
Setup Matt Pocock SkillsScaffold the per-repo configuration that Matt Pocock’s engineering agent skills rely on so they understand the issue tracker, triage labels, and domain documentation la495k196k
Lark AppsConnect Claude, Cursor or custom agents directly to Lark (Feishu) for messaging, document automation, approval workflows and enterprise data access.402k
CavemanSlash token usage by roughly 75% while keeping every technical detail intact when working with Claude Code, Cursor or similar agents.391k94.7k
Lark Skill MakerQuickly turn any Lark/Feishu OpenAPI call or multi-step workflow into a reusable agent skill with its own SKILL.md.386k16k
Running Claude Code Via Litellm CopilotRun Claude Code at a fraction of the cost by routing requests through LiteLLM to the GitHub Copilot Chat API.270k72
Codex PetGenerate a complete Codex Pet spritesheet and metadata from one reference image without needing an OpenAI key or Codex Pro.246k10