clownnvd/agent-orchestrator
agent-orchestrator/SKILL.md
--- name: agent-orchestrator description: Build multi-agent systems where an orchestrator spawns specialist worker agents in parallel waves. Use when: complex projects need parallel execution, tasks require different skills/tools, work can be decomposed into independent subtasks. triggers: ["orchestrator", "multi-agent", "spawn agents", "parallel agents", "agent workers", "divide tasks", "agent pipeline"] --- # Agent Orchestrator Orchestrator → Worker Agents pattern. Decomposes complex tasks →
tools
Updated Apr 4, 2026
$ install --global
skillsauthnpx skillsauth add clownnvd/claude-code-skills agent-orchestratorInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
3
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
VirusTotalMulti-engine malware detection
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: Apr 24, 2026, 7:48 AM4.6s1 file scanned
SKILL.md
- name:
- agent-orchestrator
- description:
- Build multi-agent systems where an orchestrator spawns specialist worker agents in parallel waves. Use when: complex projects need parallel execution, tasks require different skills/tools, work can be decomposed into independent subtasks.
- triggers:
- ["orchestrator", "multi-agent", "spawn agents", "parallel agents", "agent workers", "divide tasks", "agent pipeline"]
Agent Orchestrator
Orchestrator → Worker Agents pattern. Decomposes complex tasks → spawns specialist agents in parallel waves → collects results.
Quick Start
# Run orchestrator for a project
python scripts/orchestrator.py "Build complete SaaS app with auth, UI, and API"
# With custom waves (JSON plan file)
python scripts/orchestrator.py --plan .tmp/agents/plan.json
Architecture
User task
↓
[Orchestrator] — reads CLAUDE.md + project context
↓ plans dependency graph
Wave 1 (parallel): [db-agent] [ui-agent] ← no deps
Wave 2 (parallel): [auth-agent] ← needs db-agent
Wave 3 (sequential): [api-agent] [test-agent] ← needs all
↓
.tmp/agents/{name}/output.md ← shared memory between agents
How Agents Communicate
Agents are isolated processes. They share state via files only:
| File | Purpose |
|------|---------|
| .tmp/agents/plan.json | Orchestrator plan (waves + tasks) |
| .tmp/agents/{name}/output.md | Each agent's output |
| .tmp/agents/{name}/files.json | Files created by agent |
| .tmp/agents/summary.md | Final orchestrator summary |
Spawning Agents
Each worker is a claude --print subprocess with a skill loaded:
# Spawn with skill
spawn_agent("ui-agent", task="Build sidebar", skill="claude-ui")
# Spawn without skill
spawn_agent("cleanup-agent", task="Remove unused imports")
# Parallel wave
run_wave(["db-agent", "ui-agent"]) # Both run simultaneously
# Sequential
run_wave(["auth-agent"]) # Waits for previous wave
Agent Types (Common)
| Agent | Skill | Responsibility | |-------|-------|---------------| | db-agent | database | Schema, migrations, seed data | | auth-agent | auth | Login, OAuth, sessions | | ui-agent | claude-ui / ultimateuiux | Components, pages | | api-agent | api | Routes, middleware, validation | | test-agent | — | Test cases, validation | | infra-agent | infrastructure | CI/CD, Docker, env |
Scripts
scripts/orchestrator.py— Main orchestration enginescripts/spawn_agent.py— Single agent runner (used internally)
References
references/patterns.md— Execution patterns, edge cases, anti-patterns
Related Skills
clownnvd/zustand-pro
tools
VerifiedTrustedCommunity
Zustand v5 state management for Next.js 16. Store patterns, middleware (persist/immer/devtools), SSR hydration, CV editor multi-step wizard, 20 documented errors. Triggers: zustand, store, state management, useState replacement, global state, persist, immer.
SKILL.mdUpdated Apr 4, 2026
clownnvd/vercel-react-best-practices
development
VerifiedTrustedCommunity
React and Next.js performance optimization guidelines from Vercel Engineering. This skill should be used when writing, reviewing, or refactoring React/Next.js code to ensure optimal performance patterns. Triggers on tasks involving React components, Next.js pages, data fetching, bundle optimization, or performance improvements.
SKILL.mdUpdated Apr 4, 2026
clownnvd/ultimateuiux
development
VerifiedTrustedCommunity
Ultimate UI/UX design intelligence with real app flow knowledge. 93 styles, 121 palettes, 81 font pairings, 35 charts, 79 components, 62 animations, 65 WCAG criteria, 46 responsive patterns, 46 dark mode rules, 60 design tokens, 13 stacks. PLUS: Claude.ai full UI blueprint (19 flows, all screens), PageFlows app patterns. Actions: plan, build, create, design, implement, review, fix, improve, optimize, enhance, refactor, check, clone, recreate, rebuild. Styles: glassmorphism, brutalism, neumorphism, bento, dark mode, view transitions, scroll-driven, container queries, AI-native, liquid glass, neo-minimalism, mesh gradient, geometric abstraction. Topics: color, accessibility, animation, layout, typography, spacing, shadow, gradient, responsive, dark mode, WCAG 2.2, design tokens, components, spring physics, kinetic typography, container queries, popover API, semantic tokens. Apps: claude.ai, ChatGPT-style, AI chat UI, SaaS dashboard.
SKILL.mdUpdated Apr 4, 2026
clownnvd/ui
development
VerifiedTrustedCommunity
--- name: ui description: UI quality system. 4 modes: research (design brief), score (10-category audit), fix (auto-fix from scorecard), pipeline (end-to-end chain). license: Complete terms in LICENSE.txt --- # UI Quality System One skill, 4 modes. Research real products, score UI quality, fix issues, or run the full pipeline. ## Modes | Mode | Use When | Workflow | |------|----------|---------| | **research** | Before building any page | Extract tokens → Search → Fetch → Design Brief | | **
SKILL.mdUpdated Apr 4, 2026