Qdrant Skills
Official agent skills for Qdrant vector search, from scaling to search quality.
About
Qdrant Skills is the official skills repository from Qdrant, the open-source vector database company. It is organized as hub-and-leaf skills covering scaling (latency, data volume, QPS), performance optimization (indexing, memory, search speed), search quality (diagnosis, hybrid search strategies, relevance feedback), monitoring and debugging, client SDKs, deployment options, edge, multitenancy, model migration, and version upgrades.
The skills follow the Agent Skills standard with YAML frontmatter and trigger phrases, are Apache-2.0 licensed, and are published to a browsable site alongside the repo. They encode Qdrant-specific operational knowledge (HNSW tuning, quantization, hybrid fusion) that generic agent knowledge lacks. The repo is actively maintained and vendored by downstream projects at pinned SHAs, which is a reasonable freshness signal but no substitute for checking the docs on version-sensitive settings.
What it can do
- Scaling and performance optimization skills
- Search quality diagnosis and tuning
- Deployment, monitoring, and migration guides
Setup
npx skills add qdrant/skills, or vendor the skills directory at a pinned SHA.
Prerequisites
A Qdrant instance for live operations.
Use cases
Works with
Trust score — 8/10
Scored 8/10 (verified tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 77; Trending score 76/100 from researcher assessment.
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.