/ 디렉터리 / 플레이그라운드 / Qdrant
← 모든 서버 ↗ GitHub
● 공식 qdrant 🔑 본인 키 필요

Qdrant

제작: qdrant · qdrant/mcp-server-qdrant

Give Claude durable vector memory — store, recall, and semantically search text with a minimal, opinionated Qdrant-backed MCP.

The official Qdrant MCP turns any Qdrant instance (cloud or self-hosted) into a simple semantic memory store with just two tools: qdrant-store and qdrant-find. Perfect for giving agents long-term memory, building a personal knowledge base, or prototyping RAG without writing embedding glue code.

▶ 데모 보기 📦 설치 💡 사용 사례

왜 쓰나요

핵심 기능

라이브 데모

실제 사용 모습

qdrant.replay ▶ 준비됨
0/0

설치

클라이언트 선택

~/Library/Application Support/Claude/claude_desktop_config.json  · Windows: %APPDATA%\Claude\claude_desktop_config.json
{
  "mcpServers": {
    "qdrant": {
      "command": "uvx",
      "args": [
        "mcp-server-qdrant"
      ]
    }
  }
}

Claude Desktop → Settings → Developer → Edit Config 열기. 저장 후 앱 재시작.

~/.cursor/mcp.json · .cursor/mcp.json
{
  "mcpServers": {
    "qdrant": {
      "command": "uvx",
      "args": [
        "mcp-server-qdrant"
      ]
    }
  }
}

Cursor는 Claude Desktop과 동일한 mcpServers 스키마 사용. 프로젝트 설정이 전역보다 우선.

VS Code → Cline → MCP Servers → Edit
{
  "mcpServers": {
    "qdrant": {
      "command": "uvx",
      "args": [
        "mcp-server-qdrant"
      ]
    }
  }
}

Cline 사이드바의 MCP Servers 아이콘 클릭 후 "Edit Configuration" 선택.

~/.codeium/windsurf/mcp_config.json
{
  "mcpServers": {
    "qdrant": {
      "command": "uvx",
      "args": [
        "mcp-server-qdrant"
      ]
    }
  }
}

Claude Desktop과 같은 형식. Windsurf 재시작 후 적용.

~/.continue/config.json
{
  "mcpServers": [
    {
      "name": "qdrant",
      "command": "uvx",
      "args": [
        "mcp-server-qdrant"
      ]
    }
  ]
}

Continue는 맵이 아닌 서버 오브젝트 배열 사용.

~/.config/zed/settings.json
{
  "context_servers": {
    "qdrant": {
      "command": {
        "path": "uvx",
        "args": [
          "mcp-server-qdrant"
        ]
      }
    }
  }
}

context_servers에 추가. 저장 시 Zed가 핫 리로드.

claude mcp add qdrant -- uvx mcp-server-qdrant

한 줄 명령. claude mcp list로 확인, claude mcp remove로 제거.

사용 사례

실전 활용법: Qdrant

Give a Claude agent persistent memory across sessions

👤 Builders making personal assistants or internal copilots ⏱ ~15 min beginner

언제 쓸까: You want Claude to remember user preferences, past decisions, or ongoing projects even after the chat ends.

사전 조건
  • Running Qdrant (local Docker or cloud) — docker run -p 6333:6333 qdrant/qdrant OR a Qdrant Cloud cluster URL + API key
  • COLLECTION_NAME env var set — Any string, e.g. claude_memory
흐름
  1. Teach it to store important facts
    Whenever I tell you something important about a project (deadlines, stakeholders, decisions), store it with qdrant-store, metadata {project, category}.✓ 복사됨
    → Claude starts echoing 'stored' for durable facts
  2. Verify recall works
    What do you remember about project 'atlas'? Use qdrant-find with a query like 'project atlas decisions'.✓ 복사됨
    → Relevant prior messages returned with scores
  3. Curate and forget
    Search for anything about project 'atlas' that's more than 90 days old or marked obsolete, and delete those entries.✓ 복사됨
    → List of pruned items with confirmation

결과: An assistant that actually remembers what you told it last week — scoped per-project, prunable.

함정
  • Storing every message bloats the collection and degrades recall quality — Only store explicit facts/decisions, not chit-chat. Make the 'store or not' decision part of the system prompt.
  • Collection created with wrong vector size after switching embedding models — Qdrant rejects mismatched vectors — drop and recreate the collection when you change EMBEDDING_MODEL
함께 쓰기: filesystem · notion

Build a lightweight RAG over a docs folder

👤 Devs who want RAG without a framework ⏱ ~30 min intermediate

언제 쓸까: You have 50–5000 Markdown files and want Claude to answer questions against them, with citations.

사전 조건
  • Docs on disk as Markdown — Any folder of .md files
흐름
  1. Chunk and store the docs
    Read every .md under /docs. Split into ~500-token chunks on heading boundaries. For each chunk, call qdrant-store with the text and metadata {source_path, heading}.✓ 복사됨
    → N chunks stored, one per section
  2. Query with a user question
    User asks: 'How do I rotate API keys?' Use qdrant-find to pull the top 5 most relevant chunks. Cite source_path in your answer.✓ 복사됨
    → Answer with inline [source_path] citations
  3. Measure retrieval quality
    For these 10 eval questions [list], which of the expected source paths appear in the top-5 retrieval? Report recall@5.✓ 복사됨
    → A retrieval-quality score you can improve iteratively

결과: A working RAG loop you can tune chunk size and k until quality is acceptable.

함정
  • Too-large chunks dilute the embedding and kill recall — Keep chunks under ~1000 tokens; split by headings first, then by token count as a fallback
  • Updating a doc doesn't remove old chunks — answers become stale — Use a deterministic point id (hash of source_path+heading) so upserts replace rather than duplicate
함께 쓰기: filesystem · firecrawl

Deduplicate a messy list of tickets, leads, or FAQs semantically

👤 Ops teams with a CSV of near-duplicates ⏱ ~25 min intermediate

언제 쓸까: Exact-match dedup misses things like 'reset password' vs 'how do I change my password' — you need semantic similarity.

흐름
  1. Store each item with its row id as metadata
    Read rows.csv. For each row, qdrant-store with information=<text> and metadata={row_id: <id>}.✓ 복사됨
    → N points stored
  2. Cluster by similarity
    For each row, query qdrant-find for its top-5 neighbors with score > 0.85. Output groups of row_ids that are mutually near.✓ 복사됨
    → Duplicate groups printed
  3. Pick canonical + mark rest as duplicates
    For each group, pick the longest/most-informative row as canonical. Output a CSV {row_id, canonical_id}.✓ 복사됨
    → Dedup map ready for the source system

결과: A dedup mapping CSV with confidence scores, reviewable by a human before applying.

함정
  • Similarity threshold is domain-specific — 0.85 may be too lenient or too strict — Hand-label 20 pairs first, then pick the threshold that best separates dup from non-dup
함께 쓰기: postgres · filesystem

Searchable meeting-notes memory

👤 Managers / ICs drowning in Notion/Obsidian notes ⏱ ~20 min beginner

언제 쓸까: You take weekly notes but can never find the one where a specific decision was made.

사전 조건
  • Folder of meeting notes — Any text or markdown files
흐름
  1. Index existing notes
    Walk /meetings/**/*.md. For each, qdrant-store the body with metadata {date, attendees, project}.✓ 복사됨
    → All notes indexed with dates
  2. Recall decisions
    Find every note where we discussed 'pricing for enterprise tier'. Show me the date and a 2-line summary of each.✓ 복사됨
    → Ranked list of matching meetings
  3. Keep it fresh
    Add today's note <paste>, then tell me which past notes most likely contradict or update decisions in today's.✓ 복사됨
    → Contradiction check via semantic neighbors

결과: A semantic index over your notes you can keep updating weekly.

함정
  • Mixing personal + work notes in one collection leaks scope — Use separate collections or enforce a scope metadata filter on every find
함께 쓰기: filesystem · notion

조합

다른 MCP와 조합해 10배 효율

qdrant + filesystem

Index a local docs folder then answer questions with citations

Index every .md under /docs into Qdrant, then answer: 'how does our auth flow work?' with citations to the original file paths.✓ 복사됨
qdrant + firecrawl

Crawl a site and build a searchable knowledge base

Crawl docs.mycompany.com with Firecrawl, store each page in Qdrant collection company_docs.✓ 복사됨
qdrant + postgres

Semantic search over unstructured columns in a relational DB

SELECT id, body FROM support_tickets created in the last 30 days, embed each body into Qdrant with metadata {ticket_id}, then let me search them by meaning.✓ 복사됨

도구

이 MCP가 노출하는 것

도구입력언제 호출비용
qdrant-store information: str, metadata?: object Persist a fact, chunk, or note for later semantic recall free (local embedding)
qdrant-find query: str, limit?: int Retrieve semantically similar entries to answer a question or deduplicate free

비용 및 제한

운영 비용

API 쿼터
Self-hosted: unlimited. Qdrant Cloud: depends on cluster size.
호출당 토큰
Store: ~100 tokens overhead per call. Find: ~200 tokens + result payload.
금액
Free if self-hosted. Qdrant Cloud free tier: 1GB cluster. Paid from ~$25/mo.
팁
Start with local Docker for dev; upgrade to Cloud only when you need persistence and multi-device access.

보안

권한, 시크릿, 파급범위

자격 증명 저장: QDRANT_URL and optional QDRANT_API_KEY in env vars
데이터 외부 송신: If self-hosted: none. If Qdrant Cloud: all vectors and metadata sent to your cluster region.

문제 해결

자주 발생하는 오류와 해결

Collection does not exist / Not found

The server creates the collection on first store only if COLLECTION_NAME is set. Verify the env var and restart the MCP.

확인: curl $QDRANT_URL/collections
Vector dimension mismatch

You changed EMBEDDING_MODEL without dropping the old collection. Drop it and start fresh (or use a new COLLECTION_NAME).

확인: curl $QDRANT_URL/collections/<name>
Connection refused on localhost:6333

Qdrant container isn't running. docker run -p 6333:6333 qdrant/qdrant and retry.

확인: curl localhost:6333/healthz
Searches return irrelevant results

Chunks may be too big or the embedding model too weak. Try FastEmbed's bge-small-en-v1.5 and chunks ≤500 tokens.

대안

Qdrant 다른 것과 비교

대안언제 쓰나단점/장점
Chroma MCPYou prefer an embedded vector DB with zero infraLess production-grade than Qdrant for heavy loads
Pinecone MCPYou're already on Pinecone and want hosted-onlyPaid from day one; more opinionated
Memory MCPYou want ultra-simple key-value memory, not semanticNo embeddings — exact recall only

더 보기

리소스

📖 GitHub에서 공식 README 읽기

🐙 열린 이슈 보기

🔍 400+ MCP 서버 및 Skills 전체 보기