Note: This is a customized fork of gotalab/skillport maintained by xenitV1.

🚢 All Your Agent Skills in One Place - Manage once, serve anywhere ⚓

To use the "Zero Install" method (Option 1), you need uv.

Check if you have it:

uv --version

If not, install it:

• Windows: powershell -c "irm https://astral.sh/uv/install.ps1 | iex"
• macOS/Linux: curl -LsSf https://astral.sh/uv/install.sh | sh

You can use SkillPort without cloning the repository.

No installation required. uvx will download and run the latest version automatically. See the Editor Configuration section below.

To install it permanently on your system:

pip install git+https://github.com/xenitV1/skillport.git

Run the server to ensure it's ready:

# If installed via pip:
python -m skillport

# If using uvx:
uvx --from git+https://github.com/xenitV1/skillport skillport

Note: Configuration is automatically handled by the included .skillportrc file.

To enable the "Brain" (Antigravity Workflows), you must register the local skills directory:

# If installed via pip:
python -m skillport add https://github.com/xenitV1/skillport/tree/main/.agent/skills

# If using uvx:
uvx --from git+https://github.com/xenitV1/skillport skillport add https://github.com/xenitV1/skillport/tree/main/.agent/skills

This project combines two powerful tools:

1. The Engine (SkillPort): Universal MCP server for loading skills.
2. The Brain (Antigravity Skills): Expert-level operational workflows.

Add this to your MCP config file (VS Code, Claude Desktop, Cursor).

Option A: Using UVX (Zero Install)

{
  "mcpServers": {
    "skillport": {
      "command": "uvx",
      "args": [
        "--from",
        "git+https://github.com/xenitV1/skillport",
        "skillport"
      ]
    }
  }
}

Option B: If Installed via PIP

{
  "mcpServers": {
    "skillport": {
      "command": "skillport",
      "args": []
    }
  }
}

# Search for skills
python -m skillport search "python"

# Add a skill
python -m skillport add hello-world