Language: 日本語English

qdrant-skills Plugin

Author
Qdrant
Category
Database
Topics
Databases & Storage · Claude Code Customization & Workflow · AI Agents & AI App Development
First seen
2026-07-09
Last confirmed
2026-07-13
Explanation last updated
2026-07-12

The explanation below is AI-generated. Please verify it against the sources.

qdrant-skills is a plugin providing a set of agent skills that encode Qdrant vector search knowledge for coding agents. According to the homepage and README, it covers scaling, performance optimization, search quality, monitoring, deployment options, an embedded 'edge' shard, embedding model migration, version upgrades, and SDK usage across Python, TypeScript, Rust, Go, .NET, and Java. The README states that skills are not documentation but answer 'when' and 'why' rather than 'how', structured like a Solutions Architect's handbook that points to the exact relevant documentation for a given problem. Skills can be used by passing the skills.qdrant.tech URL directly to an agent with no installation, or installed locally via the npx skills CLI, Claude Code plugin marketplace, Cursor, or by cloning the repo into an agent-specific skills directory. The README also notes that the project is under active development, so skill content and structure may change over time.

Overview

qdrant-skills is not a SaaS product but a plugin/skill package: a collection of markdown-based agent skills hosted at skills.qdrant.tech and in the qdrant/skills GitHub repository, designed to be consumed by coding agents (e.g., Claude Code, Cursor, OpenCode, OpenAI Codex, Pi) either via URL reference or local installation.

What you can do with qdrant-skills

  • Reference qdrant-clients-sdk for SDK setup and code examples across Python, TypeScript, Rust, Go, .NET, and Java
  • Use qdrant-scaling and its sub-skills to decide on scaling for data volume, QPS, query volume, or latency, and horizontal vs. vertical approaches
  • Use qdrant-performance-optimization and its sub-skills to diagnose and fix slow search, high memory usage, or slow indexing
  • Use qdrant-search-quality and its sub-skills to diagnose irrelevant/missing results and choose search strategies such as hybrid search or reranking
  • Use qdrant-monitoring and its sub-skills to set up Prometheus, health probes, alerting, log centralization, and debug production issues
  • Use qdrant-deployment-options to choose between local, Docker, self-hosted, Cloud, or embedded deployments
  • Use qdrant-edge for guidance on the embedded in-process shard, server sync, on-device BM25, and snapshots
  • Use qdrant-model-migration to switch embedding models without downtime
  • Use qdrant-version-upgrade for safe upgrade paths and rolling upgrades
  • Install skills via npx skills add qdrant/skills, the Claude Code plugin marketplace, Cursor Marketplace, or by cloning the repo into the relevant agent skills directory
  • Pair skills with the mcp-code-snippets or mcp-server-qdrant MCP servers for additional Qdrant context, per the README

Sources

Original description (English)

Agent skills for Qdrant vector search covering scaling, performance optimization, search quality, monitoring, deployment, model migration, version upgrades, and SDK usage across Python, TypeScript, Rust, Go, .NET, and Java.

History of qdrant-skills

Back to list