skills-manager-mcp
Acts as a personal AI development environment manager, enabling workspace bootstrapping, headless skill and bundle installation, global caching, and health diagnostics.
README
Skills Manager MCP Server & CLI (skills-manager-mcp)
A production-ready CLI tool and Model Context Protocol (MCP) server for Antigravity Desktop that acts as your personal AI development environment manager.
It features zero-config automatic first-run initialization, workspace detection, headless skill & bundle installation, global skill caching, personal skill collection merging, schema version migrations, and health diagnostics.
Installation
Install globally using standard npm or pnpm (no extra security flags or install script approvals required):
npm install -g skills-manager-mcp
Automatic First-Run Initialization
skills-manager-mcp does not rely on npm postinstall lifecycle scripts (which npm security policies often block).
Instead, the very first time you run any command:
skills-manager-mcp status
(or when Antigravity Desktop invokes any MCP tool), the tool automatically detects first-time usage and configures itself:
Skills Manager MCP first-time setup detected...
✓ Global storage initialized: C:\Users\<username>\.ai-skills
✓ Skills cache ready: C:\Users\<username>\.ai-skills\cache
✓ Antigravity MCP registered: C:\Users\<username>\.gemini\antigravity-ide\mcp.json
✓ Server executable path: C:\Users\<username>\AppData\Roaming\npm\node_modules\skills-manager-mcp\dist\index.js
Initialization complete.
Subsequent executions skip initialization instantly because all configurations are already verified and active!
CLI Commands
1. skills-manager-mcp status
Displays the status dashboard (global config presence, cache statistics, Antigravity MCP registration status, detected workspace, and installed skills). Triggers auto-initialization on first run.
2. skills-manager-mcp doctor
Runs diagnostic health checks on your installation, configuration, global cache, and MCP registration.
skills-manager-mcp doctor
Example Output:
Skills Manager Doctor
✓ dist/index.js exists
✓ Antigravity configuration exists
✓ MCP path valid
✓ Global cache available
✓ skills.config.json valid
Everything is healthy.
3. skills-manager-mcp bootstrap
Prepares the current project workspace automatically.
- Detects active workspace
- Creates
.agents/skills/ - Merges personal (
~/.ai-skills/skills.config.json) and project skills - Installs skills/bundles headlessly from global cache or remote Git repositories
- Updates version metadata tracker (
.agents/skills-manager.json)
4. skills-manager-mcp sync
Synchronizes current workspace skills with your global personal collection (~/.ai-skills/skills.config.json).
5. skills-manager-mcp remove <skill1> [skill2]
Removes one or an array of specified skills or bundles from the project workspace .agents/skills.
- Option
--from-configalso removes them fromskills.config.jsonto prevent auto-reinstallation.
# Remove single skill
skills-manager-mcp remove find-skills
# Batch remove multiple skills & bundles
skills-manager-mcp remove find-skills frontend-design mattpocock-skills --from-config
6. skills-manager-mcp setup
Re-run setup & Antigravity MCP registration manually anytime.
Configuration (~/.ai-skills/skills.config.json)
You can manage your master list of skills and bundles globally at C:\Users\<username>\.ai-skills\skills.config.json:
{
"skills": [
{
"type": "skill",
"name": "find-skills",
"repository": "https://github.com/vercel-labs/skills",
"skill": "find-skills"
},
{
"type": "skill",
"name": "frontend-design",
"repository": "https://github.com/anthropics/skills",
"skill": "frontend-design"
},
{
"type": "bundle",
"name": "mattpocock-skills",
"repository": "https://github.com/mattpocock/skills"
}
]
}
MCP Server Tools Reference
When running inside Antigravity Desktop, the following MCP tools are available:
bootstrap_project: Prepares active workspace with skills and bundles.sync_skills: Synchronizes workspace skills with personal collection.install_skills: Installs missing skills and bundles.remove_skills: Removes one or an array of specified skills/bundles from workspace.list_installed_skills: Lists installed skills and bundle details.check_missing_skills: Checks status of skills/bundles.get_workspace_info: Debugging tool for inspecting workspace resolution.
Development & Testing Commands
# Install dependencies
pnpm install
# Run unit tests
pnpm test
# Compile TypeScript (generates dist/index.js and dist/cli.js)
pnpm run build
# Create npm distribution tarball
npm pack
# Link locally for testing
npm link
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.