MCP servers and agent tools for Qdrant
AgentSidekick indexes 17 DER agent products that mention Qdrant; 16 of them have a full profile and are listed below. The most-starred tools that are their own repository are mcp-server-qdrant (1.6k stars), memory-os (1.4k stars) and qdrant-client (1.4k stars) SRC. Figures as of 2026-11-21.
Qdrant tools with a full profile
Search the directoryRanked by GitHub stars of the product's own repository. Products that live in a subfolder of a larger repository (marked “repo”) show that whole repository's stars and are ranked after them; products without a public repository come last.
| # | Product | Type | Hosting | Licence | Stars | Last push | Publisher | Save and compare |
|---|---|---|---|---|---|---|---|---|
| 1 | mcp-server-qdrantAn official Qdrant Model Context Protocol (MCP) server implementation | MCP server | — | Apache-2.0 | 1.6k | 1d ago | qdrant | |
| 2 | memory-osA 7-layer memory operating system for Hermes Agent — persistent memory with Qdrant, structured facts, fabric recall, auto-curated wiki, and surgical context injection. Runs locally, any LLM provider. | Infrastructure | — | MIT | 1.4k | 2mo ago | ClaudioDrews | |
| 3 | qdrant-clientPython client for Qdrant vector search engine | Infrastructure | — | Apache-2.0 | 1.4k | 11d ago | qdrant | |
| 4 | NornicDBNornicdb is a distributed low-latency, Graph+Vector, Temporal MVCC with all sub-ms HNSW search, graph traversal, and writes. Using Neo4j Bolt/Cypher and qdrant's gRPC means you can switch with no changes while adding intelligent features like schemas, managed embeddings, reranking+llm, GPU accel, Auto-TLP, Policy-based Memory Decay, and MCP server. | MCP server | — | MIT | 933 | today | orneryd | |
| 5 | shopkeeper-agent📊 电商数仓智能问数 AI Agent,最适合用于系统学习 LangGraph 的实战项目:基于 LangGraph、FastAPI、Qdrant、Elasticsearch、MySQL 与 React,完整实现元数据知识库、混合检索、自然语言生成 NL2SQL 生成校验、SQL 执行与流式查询展示。前后端完整代码全栈可跑,Docker 环境一键部署,配套 ai-agents-from-zero 免费教程与章节代码分支。适合系统学习大模型应用、数据分析 Agent 和企业级 AI 工程落地。 | Infrastructure | — | MIT | 548 | 15d ago | didilili | |
| 6 | qdrant-skillsAgent 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. | Plugin | Local | Apache-2.0 | 271 | today | qdrant | |
| 7 | MCP AutoMemGraph-vector memory for AI assistants using FalkorDB and Qdrant | MCP server | Remote and local | MIT | 67 | 3d ago | verygoodplugins | |
| 8 | fegisDefine AI tools in YAML with natural language schemas. All tool usage is automatically stored in Qdrant vector database, enabling semantic search, filtering, and memory retrieval across sessions. | MCP server | — | MIT | 24 | 1y ago | p-funk | |
| 9 | Qdrant MCPThe full Qdrant API as 58 MCP tools: plain-text semantic/BM25/hybrid search and a search advisor. | MCP server | Local | MIT | 2 | 6d ago | avaazquezz | |
| 10 | Qdrant MemoryPersistent scoped Qdrant memory for agents, through four explicit stdio MCP tools. | MCP server | Local | MIT | 1 | 2d ago | cnkang | |
| 11 | MCP DB ConnectMCP server for Oracle, MSSQL, PostgreSQL, MySQL, SQLite, MongoDB, and Qdrant vector search access. | MCP server | Local | MIT | 1 | 3mo ago | phatngoit | |
| 12 | Qdrant RBACRole-based access to your Qdrant vector store, via MCP. OIDC in, scoped JWTs out. | MCP server | Local | MIT | 0 | 6d ago | fidonis.de | |
| 13 | vectorsearchStore and retrieve text in a Qdrant vector database by similarity search. | MCP server | Local | MIT | 0 | 21d ago | CloudCorpRecords | |
| 14 | hunt-rag-vectorHunt vector-store / embedding-layer weaknesses in RAG pipelines (OWASP LLM08 Vector and Embedding Weaknesses) — persistent corpus poisoning that survives across sessions and users (distinct from one-shot indirect prompt injection, which is owned by hunt-llm-ai), cross-tenant vector-database IDOR (unauthenticated or unscoped queries against Pinecone/Weaviate/Chroma/Milvus/Qdrant/pgvector), source-text/metadata leakage in similarity-search results, and retrieval-hijack via adversarial embedding proximity ('SEO poisoning' for RAG). Targets: any app with a shared knowledge base, document upload feeding a chatbot, or a directly reachable vector-DB port. Validate: a second, clean session/account must inherit a poisoned result, or a cross-tenant artifact must be independently verifiable — confabulation is not a finding, same bar as hunt-llm-ai. Use when target is RAG-backed, exposes a vector-DB port, or lets users upload documents that other users' queries later retrieve. | Agent skill | Local | MIT | — | today | elementalsouls | |
| 15 | GPUvecCompare GPU specs and cloud rental offers, check CUDA support, and estimate Qdrant resources. | MCP server | Remote | — | — | — | abdelkareemkobo | |
| 16 | mcp-qdrantRead-only Qdrant vector DB for AI agents — collections, info, browse points, count, search. | MCP server | Local | — | — | — | infoinlet-marketplace |
Hosting across all 16 profiled: 1 remote, 8 local, 1 both, 6 not declared. DER
From the vendor's own organisation vs the communityDER
3 of 16 profiled products are published by an account matching Qdrant's own name; the other 13 are from other publishers.
| Product | Publisher | Why it counts as the vendor's | Stars |
|---|---|---|---|
| mcp-server-qdrant | qdrant | Matched on publisher name “qdrant” | 1.6k |
| qdrant-client | qdrant | Matched on publisher name “qdrant” | 1.4k |
| qdrant-skills | qdrant | Matched on publisher name “qdrant” | 271 |
How this was determined: the publisher name, its GitHub owner, the repository owner and (for MCP Registry DNS namespaces) the domain's main label were compared with qdrant
. A vendor publishing under a differently named account (for example a parent company) is counted as community; a same-named account is not proof of official status.
Product-type mix
Which forms the tools take. Indexed covers every matching record; profiled only those with a page.
| Type | Indexed | Share | Profiled |
|---|---|---|---|
| MCP servers | 12 | 70.6% | 11 |
| Infrastructure | 3 | 17.6% | 3 |
| Plugins | 1 | 5.9% | 1 |
| Agent skills | 1 | 5.9% | 1 |
What these tools do
Most common capabilities among the profiled tools, matched from their descriptions by keyword rules.
Capabilities of Qdrant toolsDER
Profiled products mentioning Qdrant, by capability, top 10. A product can have several. As of 2026-11-21.
Show the data as a table
| profiled products | |
|---|---|
| Vector Databases | 16 |
| Vector Storage | 11 |
| Retrieval-Augmented Generation | 6 |
| Semantic Search | 3 |
| Similarity Search | 3 |
| Knowledge Bases & Wikis | 2 |
| Embeddings | 2 |
| Knowledge Graphs | 2 |
| SQL Databases | 2 |
| NoSQL Databases | 2 |
Demand signal
How often developers discuss this integration alongside MCP in public GitHub issues.
1,191 SRC public GitHub issues contain the phrase Qdrant mcp
(search run 2026-11-21). Run the same search on GitHub.
This measures developer discussion, not buyers or revenue: an issue may be a bug report, a feature request or a passing mention. The count is GitHub's own estimate for the query "Qdrant mcp" in:title,body is:issue.
Segment statisticsDER
All 17 indexed products, not only profiled ones. Calculated 2026-11-21.
- Launches
- 6 in the last 90 days, 3 in the 90 days before (+100%)DER
- Products with a repository
- 13 of 17 have whole-repository star countsDER
- Combined stars
- 6,200DER
- Median stars
- 67DER
- Largest project's share
- 25% of all stars in the segmentDER
- Concentration (HHI)
- 0.197, as concentrated as about 5 equally starred projectsDER
- Open-source share
- 82%DER
- Label
- Lagging the catalogueDER
How the label is set: size is this integration's percentile among all 111 integrations by indexed products (80th or above is crowded, 30th or below is thin). Momentum compares the ratio of launches in the last 90 days to the 90 before with the same ratio for the whole catalogue, so a booming ecosystem does not make every segment look like it is accelerating; it needs at least 5 launches across both windows. A segment can launch more products than last quarter and still be labelled slowing if the catalogue grew faster.
Vendor API
Qdrant has a public API CUR. API docs (as recorded) The link has not been checked since it was recorded.
Qdrant is not recorded as primarily a paid business product.
Opportunity score
Better agent tooling for Qdrant scores 49.1 HEU of 100 (demand component 59, gap 39; competition high, confidence medium; calculated 2026-11-21). Scores rank integrations against each other; they do not predict revenue.
How products are matched, and the limits
A product is counted when its name, description or repository topics contain one of these aliases (case-insensitive, whole words): qdrant
. A product named exactly “Qdrant” is skipped, since it is the service itself.
Which alias matched, across the 17 indexed records: qdrant
17.
Keyword matching over-counts tools that only mention Qdrant in passing (for example as where their code is hosted) and misses tools that integrate with it without naming it. Treat the counts as an upper bound on dedicated tools and a lower bound on everything that touches Qdrant.
Sources
- GitHub issue search: github.com (external)checked
- API documentation (as recorded, unverified): qdrant.tech (external)check date not recorded
Product records come from the official MCP Registry, GitHub, Open VSX and plugin marketplaces; each profile lists its own sources. Methodology
Page data last updated .