QdrantSkillsMCP 1.0.0

QdrantSkillsMCP

A .NET 10 MCP (Model Context Protocol) server for vector-based skill storage and retrieval using Qdrant. Enables AI agents (Claude Code, Copilot, Codex, etc.) to semantically search, load, and manage skills via MCP tools.

Install

# .NET 10+ (ephemeral, no install)
dnx QdrantSkillsMCP --console help

# Or install globally
dotnet tool install -g QdrantSkillsMCP
qdrant-skills-mcp --console help

Quick Start

# Initialize config with local Qdrant preset
qdrant-skills-mcp --config init

# Auto-configure your AI agent (Claude, Copilot, Codex, etc.)
qdrant-skills-mcp --setup

# Search skills via CLI
qdrant-skills-mcp --console search "authentication"

# Interactive REPL
qdrant-skills-mcp --console

Configuration

# Show all config with sources
qdrant-skills-mcp --config show

# Set remote Qdrant host
qdrant-skills-mcp --config set QdrantHost=my-qdrant.example.com
qdrant-skills-mcp --config set UseTls=true
qdrant-skills-mcp --config set QdrantApiKey=your-api-key

# Switch between profiles
qdrant-skills-mcp --config use cloud

# Validate connection
qdrant-skills-mcp --config validate

# Generate env var template for your shell
qdrant-skills-mcp --config env

MCP Tools

When running as an MCP server (default mode), exposes these tools to agents:

Embedding Providers

Configurable via --config set EmbeddingProvider=<provider>:

• LocalONNX (default) — all-MiniLM-L6-v2, runs locally, no API key needed
• OpenAI — text-embedding-3-small/large
• Ollama — any Ollama embedding model
• AzureOpenAI — Azure-hosted OpenAI embeddings

Development

Requires .NET 10 SDK and Docker (for Qdrant via Aspire).

# Run with Aspire (starts Qdrant automatically)
dotnet run --project src/QdrantSkillsMCP.AppHost

# Run tests
dotnet test

License

MIT