Skills for designing and evaluating multi-agent systems: orchestrator/worker decomposition, output quality review, and self-improving evaluator/optimizer loops.
Design and run a self-improving evaluator/optimizer loop. Use this skill when asked to "set up an eval loop", "build an optimizer", "improve this output iteratively", or "create a generate-evaluate-improve cycle" for any agent output.
Decompose a task into an orchestrator and worker architecture. Use this skill when asked to "plan an agent", "design a multi-agent system", "break down this task for agents", or any request that involves delegating work across multiple Claude instances or tool-calling pipelines.
Evaluate the output quality of an agent or pipeline run. Use this skill when asked to "review this agent output", "score this result", "evaluate agent quality", or "suggest improvements" to an agent's response or pipeline output.
A plugin library and development framework for Claude Code. This repo serves two purposes:
/plugin install.# Add the marketplace
/plugin marketplace add ats-kinoshita-iso/agent-workshop
# Browse and install plugins
/plugin install planning@agent-workshop
Browse cookbook/ and copy what you need into your project:
cookbook/claude-md/ — CLAUDE.md templates by project typecookbook/hooks/ — Reusable hook recipes for .claude/settings.jsoncookbook/mcp/ — MCP server configurations for common integrations.claude-plugin/ # Marketplace definition (marketplace.json)
plugins/ # Stable, packaged Claude Code plugins
cookbook/ # Golden baseline configs (copy into your projects)
claude-md/ # CLAUDE.md templates
hooks/ # Hook recipes
mcp/ # MCP server configs
tools/ # Development & validation tooling
tests/ # Plugin validation gates
1. Develop in .claude/ → Iterate locally with Claude Code
2. Validate with test suite → uv run pytest
3. Package as plugin → Create plugin.json + SKILL.md in plugins/
4. Auto-register in marketplace → marketplace_gen.py updates marketplace.json
5. Users install from here → /plugin marketplace add ats-kinoshita-iso/agent-workshop
| Plugin | Description | Version |
|---|---|---|
| code-quality-gate | Unified quality orchestrator (lint, format, typecheck, test) | 1.0.0 |
| context-sync | Keeps CLAUDE.md files in sync with the codebase | 1.0.0 |
| plan-manager | Plan lifecycle management with gate tracking and archival | 1.0.0 |
| planning | Phased implementation plans with gates and tests | 2.0.0 |
| test-quality | Test generation, auditing, and knowledge extraction | 1.0.0 |
| workspace-clean | Workspace hygiene checks and cleanup | 1.0.0 |
| Plugin | Skills | License |
|---|---|---|
| anthropic-document-skills | docx, pdf, pptx, xlsx | Source-available |
| anthropic-creative-skills | algorithmic-art, brand-guidelines, canvas-design, frontend-design, slack-gif-creator, theme-factory | Apache 2.0 |
| anthropic-dev-skills | claude-api, mcp-builder, skill-creator, web-artifacts-builder, webapp-testing | Apache 2.0 |
| anthropic-enterprise-skills | doc-coauthoring, internal-comms | Apache 2.0 |
uvbunuv syncuv run pytestuv run ruff check .uv run ruff format .uv run mypy .bun installbun testbunx biome check --write ..claude/ (skills, hooks, agents, etc.)plugins/<your-plugin>/.claude-plugin/plugin.json with name, description, version, keywordsuv run pytest to validate structureuv run python tools/marketplace_gen.py to update the marketplace catalogOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
npx claudepluginhub ats-kinoshita-iso/agent-workshop --plugin agent-patternstrine-eval: Planner-Generator-Evaluator harness for eval-driven development across web/RAG/CLI/API projects, eval-harness methodology audits (meta layer), and harness-build agent-runtime conformance (runtime layer)
Analyzes and rewrites prompts for Claude Code, applying structured prompt engineering patterns to produce clearer, more effective instructions.
Anthropic's official development skills for Claude API integration, MCP server building, skill creation, web artifact building, and browser-based testing.
Observability skills for designing logging and tracing strategies, instrumenting existing code with structured log points, and analyzing trace logs to diagnose production issues.
Evaluation framework skills for designing scoring rubrics, running structured evaluations on LLM outputs, and comparing candidate outputs to recommend a winner.
Automatically detects and loads AGENTS.md files to provide agent-specific instructions alongside CLAUDE.md. Enables specialized agent behaviors without manual intervention.
Dynamic orchestrator plugin — 200+ agents, 200+ skills, hooks, and MCP servers. Upstream sources linked via git submodules (agency-agents, everything-claude-code, oh-my-claudecode, gstack, superpowers)
Battle-tested configuration system for Claude Code agents: 28 architectural principles (incl. P-28 Feature-Layer Architecture for project-knowledge-as-a-navigable-tree), 19 safety hooks, 23 skills (incl. /layer-new + /feature-new for ULTRAPACK-style feature narratives), 16 alternative comparisons, 12 templates (incl. extended kb-skeleton with docs/layers/), diagnostic scripts (incl. build_kb_graph.py + validate_kb_links.py). Pairs with mclaude for multi-session coordination.
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Complete collection of battle-tested Claude Code configs from an Anthropic hackathon winner - agents, skills, hooks, and rules evolved over 10+ months of intensive daily use
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.