By chkim-su
Multi-command workflow system for creating Claude Code plugin components with daemon-owned policy and explicit phase transitions
Create plugin component files using multi-agent workflow
Analyze plugin component quality using multi-agent workflow
Plan component structure using multi-agent workflow
Verify plugin components using multi-agent workflow
Dispatcher - classifies intent and routes to appropriate workflow
General-purpose analysis agent. Validates schemas, scores quality, checks patterns. Used by /assist:verify and /assist:health-check workflows.
General-purpose discovery agent. Finds plugin locations, existing components, and patterns. Used by /assist:create, /assist:verify, and /assist:health-check workflows.
Executes file creation based on a plan. Use this agent after semantic phase to generate actual component files.
General-purpose reporting agent. Aggregates results, detects patterns, produces final reports. Used by /assist:health-check workflow.
Routes user requests to the appropriate workflow. Use this agent when starting a new /assist:wizard workflow to classify the user's intent.
This skill provides exact templates and file locations for creating Claude Code plugin components.
Aggregation phase for /assist:health-check command - produces final report
Analysis phase for /assist:health-check command - scores each component
This skill provides criteria and metrics for analyzing the health and suitability of Claude Code plugin components.
Discovery phase for /assist:health-check command - finds components for health analysis
Uses power tools
Uses Bash, Write, or Edit tools
Own 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.
Workflow engine-based system for creating Claude Code plugin components with structured phases and state management.
Forge3 provides a structured workflow for creating plugin components (skills, agents, commands, hooks) with:
# Install dependencies
pip install -r requirements.txt
# Start workflow daemon
workflow-daemon
# Create a new component
/assist create a skill for generating commit messages
# Verify a component
/verify skills/my-skill
forge3/
├── hooks/ # Event hooks
│ ├── hooks.json # Hook configuration
│ ├── router_hook.py # UserPromptSubmit handler
│ ├── phase_hook.py # PreToolUse enforcement
│ └── stop_hook.py # Stop prevention
├── agents/ # Subagents
│ ├── router-agent.md # Intent classification
│ ├── semantic-agent.md # Structure planning
│ ├── execute-agent.md # Implementation
│ └── verify-agent.md # Validation
├── skills/ # Skill definitions
├── commands/ # Slash commands
└── scripts/ # Utilities
csc/
└── workflow-daemon/ # Workflow engine (SSOT daemon)
| Phase | Agent | Purpose |
|---|---|---|
| Router | router-agent | Classify intent |
| Semantic | semantic-agent | Plan structure |
| Execute | execute-agent | Create files |
| Verify | verify-agent | Validate result |
| Endpoint | Method | Purpose |
|---|---|---|
/workflow/init | POST | Initialize workflow |
/workflow/status | GET | Query current state |
/workflow/transition | POST | Validate & transition |
/workflow/current | GET | Get context |
/agent/invoke | POST | Record agent invocation |
/agent/complete | POST | Mark agent done |
/sse/events | GET | SSE subscription |
# Monitor workflow events
python scripts/workflow_monitor.py
# Check current status
python scripts/workflow_monitor.py --status
# Filter by workflow ID
python scripts/workflow_monitor.py --workflow-id <id>
# Start server in development mode
cd forge3
workflow-daemon
# Test endpoints
curl http://127.0.0.1:8766/health
curl http://127.0.0.1:8766/workflow/status
Unit Test - workflow daemon endpoints
curl http://127.0.0.1:8766/workflow/status
Hook Test - Hook communication
echo '{"prompt":"/assist create skill"}' | python3 hooks/router_hook.py
E2E Test - Full workflow
/assist create a hello skillMIT
npx claudepluginhub chkim-su/forge3Strict migration plugin with bootstrap requirement: /bootstrap must run first, manual workarounds blocked, enforcement scripts must exist for /migrate to work
Oracle-powered autonomous development pipelines with Gemini-Claude orchestration
Complete plugin development toolkit for creating, refactoring, and validating Claude Code plugins and agents. Use when creating new plugins/skills/agents, refactoring existing plugins/skills, validating frontmatter, or restructuring plugin components. Includes specialized agents for assessment, planning, execution, and validation workflows.
Harness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
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.
Access thousands of AI prompts and skills directly in your AI coding assistant. Search prompts, discover skills, save your own, and improve prompts with AI.
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
Complete developer toolkit for Claude Code