Agent skill
plugin-forge
Create and manage Claude Code plugins with proper structure, manifests, and marketplace integration. Use when creating plugins for a marketplace, adding plugin components (commands, agents, hooks), bumping plugin versions, or working with plugin.json/marketplace.json manifests.
Install this agent skill to your Project
npx add-skill https://github.com/davila7/claude-code-templates/tree/main/cli-tool/components/skills/development/plugin-forge
SKILL.md
CC Plugin Forge
Purpose
Build and manage Claude Code plugins with correct structure, manifests, and marketplace integration. Includes workflows, automation scripts, and reference docs.
When to Use
- Creating new plugins for a marketplace
- Adding/modifying plugin components (commands, skills, agents, hooks)
- Updating plugin versions
- Working with plugin or marketplace manifests
- Setting up local plugin testing
- Publishing plugins
Getting Started
Create New Plugin
Use create_plugin.py to generate plugin structure:
python scripts/create_plugin.py plugin-name \
--marketplace-root /path/to/marketplace \
--author-name "Your Name" \
--author-email "your.email@example.com" \
--description "Plugin description" \
--keywords "keyword1,keyword2" \
--category "productivity"
This automatically:
- Creates plugin directory structure
- Generates
plugin.jsonmanifest - Creates README template
- Updates
marketplace.json
Bump Version
Use bump_version.py to update versions in both manifests:
python scripts/bump_version.py plugin-name major|minor|patch \
--marketplace-root /path/to/marketplace
Semantic versioning:
- major: Breaking changes (1.0.0 → 2.0.0)
- minor: New features, refactoring (1.0.0 → 1.1.0)
- patch: Bug fixes, docs (1.0.0 → 1.0.1)
Development Workflow
1. Create Structure
Manual approach (if not using script):
mkdir -p plugins/plugin-name/.claude-plugin
mkdir -p plugins/plugin-name/commands
mkdir -p plugins/plugin-name/skills
2. Plugin Manifest
File: plugins/plugin-name/.claude-plugin/plugin.json
{
"name": "plugin-name",
"version": "0.1.0",
"description": "Plugin description",
"author": {
"name": "Your Name",
"email": "your.email@example.com"
},
"keywords": ["keyword1", "keyword2"]
}
3. Register in Marketplace
Update .claude-plugin/marketplace.json:
{
"name": "plugin-name",
"source": "./plugins/plugin-name",
"description": "Plugin description",
"version": "0.1.0",
"keywords": ["keyword1", "keyword2"],
"category": "productivity"
}
4. Add Components
Create in respective directories:
| Component | Location | Format |
|---|---|---|
| Commands | commands/ |
Markdown with frontmatter |
| Skills | skills/<name>/ |
Directory with SKILL.md |
| Agents | agents/ |
Markdown definitions |
| Hooks | hooks/hooks.json |
Event handlers |
| MCP Servers | .mcp.json |
External integrations |
5. Local Testing
# Add marketplace
/plugin marketplace add /path/to/marketplace-root
# Install plugin
/plugin install plugin-name@marketplace-name
# After changes: reinstall
/plugin uninstall plugin-name@marketplace-name
/plugin install plugin-name@marketplace-name
Plugin Patterns
Framework Plugin
For framework-specific guidance (React, Vue, etc.):
plugins/framework-name/
├── .claude-plugin/plugin.json
├── skills/
│ └── framework-name/
│ ├── SKILL.md
│ └── references/
├── commands/
│ └── prime/
│ ├── components.md
│ └── framework.md
└── README.md
Utility Plugin
For tools and commands:
plugins/utility-name/
├── .claude-plugin/plugin.json
├── commands/
│ ├── action1.md
│ └── action2.md
└── README.md
Domain Plugin
For domain-specific knowledge:
plugins/domain-name/
├── .claude-plugin/plugin.json
├── skills/
│ └── domain-name/
│ ├── SKILL.md
│ ├── references/
│ └── scripts/
└── README.md
Command Naming
Subdirectory-based namespacing with : separator:
commands/namespace/command.md→/namespace:commandcommands/simple.md→/simple
Examples:
commands/prime/vue.md→/prime:vuecommands/docs/generate.md→/docs:generate
Version Management
Important: Update version in BOTH locations:
plugins/<name>/.claude-plugin/plugin.json.claude-plugin/marketplace.json
Use bump_version.py to automate.
Git Commits
Use conventional commits:
git commit -m "feat: add new plugin"
git commit -m "fix: correct plugin manifest"
git commit -m "docs: update plugin README"
git commit -m "feat!: breaking change"
Reference Docs
Detailed documentation included:
| Reference | Content |
|---|---|
references/plugin-structure.md |
Directory structure, manifest schema, components |
references/marketplace-schema.md |
Marketplace format, plugin entries, distribution |
references/workflows.md |
Step-by-step workflows, patterns, publishing |
Scripts
| Script | Purpose |
|---|---|
scripts/create_plugin.py |
Scaffold new plugin |
scripts/bump_version.py |
Update versions |
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
verl-rl-training
Provides guidance for training LLMs with reinforcement learning using verl (Volcano Engine RL). Use when implementing RLHF, GRPO, PPO, or other RL algorithms for LLM post-training at scale with flexible infrastructure backends.
openrlhf-training
High-performance RLHF framework with Ray+vLLM acceleration. Use for PPO, GRPO, RLOO, DPO training of large models (7B-70B+). Built on Ray, vLLM, ZeRO-3. 2× faster than DeepSpeedChat with distributed architecture and GPU resource sharing.
gguf-quantization
GGUF format and llama.cpp quantization for efficient CPU/GPU inference. Use when deploying models on consumer hardware, Apple Silicon, or when needing flexible quantization from 2-8 bit without GPU requirements.
Claude Code Guide
Master guide for using Claude Code effectively. Includes configuration templates, prompting strategies "Thinking" keywords, debugging techniques, and best practices for interacting with the agent.
qdrant-vector-search
High-performance vector similarity search engine for RAG and semantic search. Use when building production RAG systems requiring fast nearest neighbor search, hybrid search with filtering, or scalable vector storage with Rust-powered performance.
behavioral-modes
AI operational modes (brainstorm, implement, debug, review, teach, ship, orchestrate). Use to adapt behavior based on task type.
Didn't find tool you were looking for?