- Conduid
- Marketplace
- #retrieval
MCP servers tagged retrieval
144 live MCP servers tagged "retrieval", ranked by trust score. Tags come from package metadata and repository topics, so this list covers servers across every category that work with retrieval.
Porta Rag MCP
RAG knowledge base как MCP-tools: поиск по документам для любого AI-клиента (Claude/Cursor/ChatGPT). FastMCP-обёртка над RAG-движком, оптимизация токенов (retrieval без LLM vs пол…
Repo Aware Code Rag
AST-aware hybrid RAG over codebases with file:line citations — FastAPI + MCP. Public demo of a private live system.
Chunktuner
Benchmark chunking strategies for your RAG corpus. Get a recommended config. CLI, Python library, and MCP server.
Bert
Long-context project memory for Claude Code / Cursor / Codex — a local MCP server with hybrid retrieval (dense + BM25 + cross-encoder rerank). Falsification-first benchmarks, hone…
Agent Context Substrate
The substrate is the asset; the model is an interchangeable client. Agent Context Substrate (ACS): model-independent memory for AI agents.
Akopia
Plug. Sync. Serve. — headless retrieval plane for your AI stack. Self-hosted RAG infra with Qdrant + Meilisearch + Redis, plugin-first, MCP-native.
Repository Memory
Citation-first memory for AI agents: one zero-dependency runtime behind Skill, audited MCP (2026-07-28), native OpenClaw tools, and CLI. Git-cited answers, evidence-based abstenti…
RBI Source MCP
Hosted MCP server for Reserve Bank of India regulatory citations. Paste a clause into Claude / ChatGPT / Cursor, get cited RBI provisions back. Retrieval-only, citation-first.
Operations Context Demo
Read-only MCP context demo with bounded retrieval, source provenance, and conservative publication controls.
Cursor Memory Project
Persistent memory, retrieval, MCP server, logging, and QA framework for Cursor-based AI-assisted development workflows.
Dewie
Agent-native retrieval over a self-enriching corpus: hybrid search + graph navigation exposed as MCP tools, so small models can punch above their weight.
dfrancislyondflabc-tech/agentic-recall
bad match as though it were one. Embedding model runs locally, nothing leaves the machine. MIT.
Apexgraph
Apex-relevance subgraph retrieval for AI agents — feed your LLM the most relevant slice of a knowledge graph, within a token budget.
OpenContextScitool
LLM context handler over an Open Knowledge Format bundle BM25 search via MCP, usage telemetry, and a health dashboard. No embeddings, no API key.
Assayer Memory MCP
A local knowledge base your agents can read — and the engines that keep it honest. Computed themes, measured links, budgeted context over MCP, a verdict funnel that drains its own…
Web Vector
Web research for AI agents in one tool call: search → read the full pages → rank → cited passages. Library, MCP server and CLI; zero API keys needed.
Memory Rag
Personal-memory RAG for an AI coding agent: BM25 + bge vectors + file-level RRF, CLI/MCP dual outlets, 18-query golden eval (18/18, MRR .880)
Agtmem
Long-term memory for coding agents that you own: Markdown on disk, a disposable SQLite index, zero runtime dependencies.
Qmd Openclaw Kit
Reusable OpenClaw integration kit for upstream QMD with bilingual docs, backend wrappers, MCP templates, and collection bootstrap scripts.
Atomic Memory Kit
Bidirectional atomic memory for agents: compile Markdown knowledge atoms into one context bundle, decompile back, and spotlight what's missing. Zero deps, no embeddings.
Tacitdb
Tacit — a provenance-first knowledge engine that self-hosts its own decision records. Rust workspace: core, keeper, MCP host, Python shell.
LLM Engineering Portfolio
Thirteen measured experiments on one RAG pipeline: retrieval evaluation, hallucination defense, prompt injection, MCP, and stateful agents. Every number produced by code in this r…
Finco
RAG over SEC EDGAR filings — hybrid BM25 + dense retrieval with rank fusion and cross-encoder reranking. +75% recall@5 over dense-only, with the eval harness that proves it.
Docs Rag Assistant
Production-grade RAG assistant over the Model Context Protocol documentation — FastAPI + MCP tools, with reranking, semantic caching, guardrails, PII masking, and hallucination de…
Ctxdb
A context database for LLMs: structured storage and hybrid retrieval (BM25 + vectors, RRF-fused, token-budgeted) over SQLite, exposed to Claude as an MCP server
Claude Code Ltm
Claude Code 的長期記憶 — 在 ~/.claude/projects 逐字記錄之上的可拋棄索引與可插拔記憶策略（全本機）
Shared Memory Server
Gives Claude Desktop, Cursor, and Claude Code one shared, persistent memory of your project's context. A decision or fact is saved once and stays available in every session and ev…
DocQA
A simple UI for a RAG (Retrieval Augmented Generation) backend API to query documents or URLs and get context based answers
Agent Tool Router
Pick the right tools for an agent task. Boring baseline, open dataset (14K traces from public agent benchmarks).
Brainmem
Agent memory gated on surprise, budgeted on attention. Beliefs carry provenance and validity intervals, so you can ask what the agent believed, when, and on what evidence.
Agent Memory Site
Compile your Markdown vault into a portable agent memory. One build, every agent: Claude, Codex, Cursor, MCP, llms.txt.
Mainrag
Self-hosted private-code context layer for AI coding agents, with MCP tools, citations, tenant boundaries and hybrid retrieval
Claim Studio
营销文案事实审校与视频合成 Agent：Qwen 经 MCP 检索资料逐句核查，缺证据即停止；5 种检索策略 Hit@5 10%→43.3%，4B/14B 对照留出集 75%→100%；148 测试、OTEL/Prometheus 可观测、人工确认绑定快照后渲染 FFmpeg 视频。 | Marketing fact-checking agent: Qwe…
Aladdin Spec
Define portable, verifiable, and versioned knowledge packs for trusted AI systems and RAG applications.
NanoJev
Build a 0.6B parallel decision model that turns states into probability distributions—no output-token decoding needed.
Claude Plugins
Claude Code plugins by roomi-fields: RTFM (open retrieval layer) + NotebookLM (citation-backed Q&A)
Vera
Local code search combining BM25, vector similarity, and cross-encoder reranking. Parses 60+ languages with tree-sitter, runs entirely offline, and returns structured results with…
Vera
Local code search combining BM25, vector similarity, and cross-encoder reranking. Parses 60+ languages with tree-sitter, runs entirely offline, and returns structured results with…
Tool Rag
RAG for AI tools: each agent gets only the relevant MCP tools its user may use. Hybrid retrieval, evaluated on 1,351 real requests (92% top-5), ToolRet (44k tools) and a permissio…
WebSearch MCP
MCP server for web search and page fetching with markdown conversion for AI model integration in applications and citations.
Exa
Exa — independent third-party profile of a public API surface, by API Evangelist. Exa is a neural web search API designed for AI agents and applications. It combines a purpose-bui…
ProjectRecall
Local hybrid-search RAG (SQLite FTS5 + Ollama embeddings, weighted RRF) over your own project notes, served to Claude Code as an MCP server
Slate
An MCP client and a retrieval MCP server over a production RAG corpus. Typed responses instead of JSON strings, so a drifting tool schema fails loudly instead of returning a confi…
Pasr
No description published.