MCP servers that ship as code — a repository or a package you install and run locally (usually over stdio). They have no hosted endpoint, so we cannot probe or grade them: instead each one shows its repository, install command, license, stars and when it was last updated.
Collected from GitHub repositories and npm, PyPI, Cargo, Go, Maven, NuGet and RubyGems packages tagged as MCP servers, the official MCP Registry, Docker's MCP catalog, ModelScope, Baidu MCP World, Cline and other public directories; one entry per repository. A server that also offers a hosted endpoint is listed under Remote instead. These entries are not counted in our reachability, verification or readiness statistics.
-
rag-pipeline-grounded-llm-genomicsgithub.com/cgjuan01/rag-pipeline-grounded-llm-genomicsupdated this year
LLM · RAG · MCP: a retrieval-augmented pipeline that answers questions over my ageing & exercise genomics research from papers and gene-level results, with cited, grounded answers, plus an MCP server exposing retrieval, citation-checking and refusal as agent tools. Open-weight & API backends.
-
climategpt-performance-enhancementgithub.com/Mkatika37/climategpt-performance-enhancementupdated this year
Grounds a climate language model in real satellite (NASA VIIRS) and water-risk (WRI Aqueduct) data via MCP tools, so answers trace back to evidence, not just the model's memory.
-
This repository provides a MCP server that retrieves information from Google Search to provide context to Large Language Models (LLMs). It uses the Google Custom Search JSON API to fetch real-time web search results, allowing AI assistants to ground their responses in up-to-date information.
Unlicense -
An LLM eval harness graded by computed ground truth, not an LLM judge. Worked example: poker, served to models as MCP tools — gpt-4o-mini goes from 26/75 to 75/75 with them.
MIT ★ 1 -
An MCP server that gives any LLM host grounded, cited retrieval over your own documents — hybrid BM25 + dense retrieval, cross-encoder-ready reranking, and built-in eval.
npm MIT -
Don't trust AI. Verify it. Open-source verification for evidence-grounded AI — deterministic grounding checks against source evidence, no LLM judge. CLI, GitHub Action, Python, MCP server, Docker.
go MIT ★ 1 -
evidence-gategithub.com/LopezDray/evidence-gateupdated this year
Stop your AI from making up facts about your own data. A pre-generation grounding gate for LLM/RAG apps — zero-dep, MIT, JS + Python, with an MCP server.
MIT -
Ground control for agent swarms: fly MAVLink drone fleets with people, LLM agents and 50 Hz onboard programs. Gamepad possession and hand-back, Gazebo-style rotor sim, LAN dogfights and leaderboard, MCP.
NOASSERTION ★ 1 -
MCP server for agentic LLM evaluation: jury scoring, agent tracing via OpenTelemetry, document-grounded QA generation, PDF reports.
pypi Apache-2.0 -
agentic-reality-mcpgithub.com/agentralabs/agentic-realityupdated this year
MCP server for AgenticReality — universal LLM access to existential grounding
cargo MIT -
Grounded Documentation for LLMs
npm MIT ★ 4 -
ucp-localgithub.com/akshay2211/universal-context-pipelineupdated this year
Universal Context Pipeline — local-first MCP server for grounding LLMs in your files
cargo Apache-2.0 -
hallumarkgithub.com/cognis-digital/hallumarkupdated this year
LLM hallucination & grounding auditor for RAG systems
NOASSERTION -
LLM analytic harness: typed Python macros over MCP, number-grounded chat, and a background forecasting engine (archived prototype)
MIT -
Agent-native bi-temporal graph store: verified temporal operators as LLM tools + trace-grounded answer verification
pypi Apache-2.0 ★ 5 -
Grounded red-blue adversarial plan refinement via MCP — tree-sitter fact extraction + LLM red/blue team analysis
Apache-2.0 -
Grounded enterprise ontology agent with Neo4j, typed retrieval, LLM tool calling, evaluation, and read-only MCP tools.
-
anchorgithub.com/myrddian/anchorupdated this year
Source-grounded chunk validation as a primitive. Spring Boot service exposing two interfaces (LLM-to-LLM and human-facing) over the same hierarchical document model.
Apache-2.0 ★ 3 -
Source-grounded MCP memory runtime and Obsidian-compatible vault template for LLM agents.
MIT ★ 76 -
mathsketch-tutorgithub.com/dadadadas111/mathsketch-tutorupdated this year
Sketch a math problem, get step-by-step feedback grounded in symbolic verification. NestJS + Firebase + mathjs CAS verifier + LLM exposed as an MCP tool.
MIT -
Payment-risk triage with logistic regression, held-out precision metrics, grounded LLM decisions, bounded RAG prompts, audit storage, MCP and Docker.
MIT -
Three-way invoice reconciliation in integer cents, exception scoring, evidence-grounded LLM routing, REST APIs, an MCP server and a Docker dashboard.
MIT -
insight-link-progithub.com/Lonishubh48/insight-link-proupdated this year
Hallucination Killer — MCP server that grounds LLM answers in live repo context and real-time docs
★ 2 -
MCP server for memory forensics with Volatility 3 — tamper-evident evidence store, DKOM/rootkit detection, and evidence-grounded LLM investigation.
Apache-2.0 -
astrodynamics-mcpgithub.com/astro-tools/astrodynamics-mcpupdated this year
Model Context Protocol server giving any LLM client (Claude, ChatGPT, Cursor, agents) authoritative astrodynamics tools — TLE/SGP4, Lambert, ground-station access, time/frame conversions, porkchop, B-plane.
pypi MIT ★ 2 -
Retrieve full text of bioRxiv, medRxiv & arXiv preprints as structured sections — to ground LLM/agent scientific reasoning, deep research, and RAG. CLI + Python + MCP.
pypi NOASSERTION ★ 4 -
Local vs. cloud LLMs triaging real Splunk alerts through the Splunk MCP Server, evaluated against manual ground truth, with a least-privilege AI identity.
-
Evidence-grounded IT support agent: Gemini function calling over an MCP-style tool server (KB search, ticket search, ticket drafts), with a 60-question eval harness, LLM judge, and OpenTelemetry tracing.
★ 1 -
Token-saving browser automation MCP server: cached actions, then a local LLM, then Claude. Ground-truth benchmark included; 0 Claude tokens across 20 real runs.
-
MCP connector that verifies a claim's grounding against reality (quotes, citations, code, repos, math) with no LLM in the verification path. Works in Claude and Gemini.
MIT ★ 2 -
LLM agent that authors and runs integration tests for a complex 3-process C payments backend through a 15-tool MCP server, grounded in a generated knowledge base, with a controlled eval measuring whether the KB helps.
MIT ★ 1 -
Zero-error LLM execution via SPRT voting. Rust library and MCP server implementing the MAKER algorithm for mathematically-grounded error correction in long-horizon AI agent tasks. Research experiment based on arXiv:2511.09030
MIT ★ 5 -
Type-safe LLM vulnerability triage on Pydantic AI + MCP. Ten typed tools (NVD, OSV, KEV, EPSS, Exploit-DB, ATT&CK, SBOM, Nmap) ground a validated TriageReport with a deterministic SSVC verdict. Red-team battery (MITRE ATLAS), reproducible scorecard, OWASP LLM + ISO 42001 mapping. Next.js frontend.
MIT ★ 2 -
perceptgithub.com/wowow4978-vorfeed/perceptupdated this year
A local service that aggregates heterogeneous sensor data into a uniform, LLM-queryable event store. An LLM can ask "what do you see now?" or "what happened last night?" and get grounded answers from real sensor history.
MIT -
Finds UAE companies with buying signals in the news, scores them, finds contact emails, drafts grounded outreach with a local LLM, and sends only after human approval. Python, Ollama, MCP server, Flask review page.
-
A protocol-driven RAG system that decouples LLM reasoning from tool execution using MCP, enabling scalable document intelligence with ChromaDB and OpenAI. Implements an MCP client–server architecture with explicit message orchestration, structured tool execution, and stateful multi-turn memory for document-grounded LLM applications
-
weather-mcp-servergithub.com/Amankhan1009/weather-mcp-serverupdated this year
Production-grade AI Weather MCP Server — combines real-time weather data with LLM reasoning to deliver intelligent, grounded insights (clothing suggestions, travel advice, activity recommendations) via the Model Context Protocol. Built with Python, FastMCP, Pydantic, Docker, and Groq.
★ 1 -
Turns Indian GST tax invoices into government INV-01 JSON. Tells you exactly what it couldn't read, and never invents a value to fill a gap. MCP server with OCR, deterministic extraction, and a grounded LLM stage.
MIT -
Ground LLM output as provenance-tracked claims, not truths. A maximal super-ontology + native RDF-1.2 logic: reasoning engine — OWL, SHACL, SSSOM, EDOAL are generated lossy projections; a correspondence calculus losslessly subsumes external vocabularies. Rust-first (pip / MCP / GTS).
AGPL-3.0 ★ 13 -
operandi-mcpupdated this year
MCP server — let any LLM agent identify an appliance and pull its grounded, safety-checked, robot-executable operation package from OPERANDI.
pypi Proprietary -
mcp-docsgithub.com/flaco-source/mcp-docsupdated this year
An MCP (Model Context Protocol) server that gives LLMs direct access to official vendor PDF documentation (starting with Texas Instruments), with a local SQLite full-text index (FTS5 + BM25) so answers can be grounded in real datasheet and TRM text.
★ 2 -
traffic_monitorgithub.com/pflaumax/traffic_monitorupdated this year
A FastAPI reverse proxy that streams traffic events through Kafka, exposes an LLM /ask endpoint that answers questions in plain language grounded in the live stats, and ships an MCP server.
-
International Education Standards MCP — CCSS, Singapore MOE, IB MYP and more. Grounding layer for LLM-powered education tools.
pypi MIT ★ 5 -
fieldraggithub.com/woongjaejung/fieldragupdated this year
Grounded RAG + agent assistant for NGS field support on Google Cloud (Gemini + pgvector/Cloud SQL + FastAPI/Cloud Run, LangGraph, MCP, LLM-judge eval)
-
CodeCheckgithub.com/SiddharthM2416/CodeCheckupdated this year
Most "chat with your codebase" tools stop at Q&A. CodeCheck goes further: it finds untested, risky code, explains why it matters, and drafts tests — all grounded in real retrieved source, not guesses. Built with tree-sitter, ChromaDB, and MCP, on a free multi-provider LLM backend with a Streamlit UI.
★ 1 -
AI-powered clinical decision support system using RAG & LLMs for disease and drug assistance, YOLOv8 for X-ray pre-screening, Google Document AI for medical report extraction and patient summaries, with LangGraph & MCP orchestrating specialized AI services through a Spring Boot and Python microservice architecture to support data-grounded clinical
MIT -
A Model Context Protocol (MCP) server that grounds LLM coding agents in up-to-date FIRST Robotics Competition (FRC) knowledge — WPILib docs and API references plus vendor ecosystems like CTRE Phoenix, REVLib, PathPlanner, PhotonVision, and AdvantageKit — so generated robot code tracks the current season instead of stale training data.
go MIT ★ 1 -
Local code search and cited answers for Project Chrono and PyChrono, grounded in Chrono's own code, demos and docs. CLI, MCP server for editors, and web app; search needs no API key, written answers use your LLM key or a local LLM. Indexes Chrono main as dated snapshots; a 10.0.0 release index is also available.
BSD-3-Clause ★ 4 -
Codebase Intelligence Engine & MCP Server — AST semantic chunking, tri-hybrid retrieval (BM25 + vectors + graph), and grounded LLM synthesis for any repository.
npm MIT -
OpenScout is a Perplexity-inspired answer engine that turns a single question into a grounded, cited response. It combines web search, retrieval, reranking, and LLM synthesis in a lightweight Streamlit Application
MIT ★ 1 -
Deterministic Git & AST change intelligence engine and Model Context Protocol (MCP) server. Detects breaking changes, maps caller blast radius, identifies doc drift, and drafts evidence-grounded release notes with optional local LLM acceleration.
Apache-2.0 ★ 13 -
Sovereign debt crisis modeling for AI agents and LLM workflows — this MCP server delivers eight academically-grounded simulation tools spanning default dynamics, contagion networks, fiscal sustainability, restructuring design, speculative attacks, and political regime transitions.
MIT -
pulse-flight-deals-mcpgithub.com/yaotsakpo/pulse-flight-deals-mcpupdated this year
MCP server that lets an AI agent judge whether a flight price is actually a deal, grounded in observed fare history. Answers 'is $X for this route a good price?' with real numbers an LLM cannot produce on its own.
npm MIT -
thorbit-knowledge-graphgithub.com/vilovietaseo/thorbit-knowledge-graphupdated this year
Build canonical entity libraries (Wikidata, Wikipedia, DBpedia, Freebase mid) from any website using TextRazor entity linking, then emit grounded schema.org JSON-LD (entity IDs locked, prose written by an LLM). Ships a CLI, an MCP server, and a deployable
npm MIT -
querywise-mcpgithub.com/kosminus/querywise-mcpupdated this year
An MCP server (and a CLI) that lets an LLM query your databases in natural language through a business semantic layer — glossary, metric definitions, data dictionary, knowledge base, and example queries — grounded against your real schema.
MIT ★ 3 -
Local RAG MCP server for markdown docs. Retrieval only (no answer LLM): ask_docs returns grounded passages plus citations that resolve to real files on disk. Index any local path or git repo, filter by layer, embed offline.
pypi MIT -
docwise-agenticgithub.com/rakmohan/docwise-agenticupdated this year
Full agentic RAG pipeline: adaptive chunking → Qdrant hybrid search (semantic + BM25 + RRF) → LangGraph ReAct agent → grounded answers with citations, streamed live. Multi-turn conversation, multi-hop reasoning, human-in-the-loop, LangSmith tracing, MCP server. LLM-agnostic — works with Claude or OpenAI.
MIT -
neruva-mcpupdated this year
Neruva MCP -- proprietary memory + grounding substrate for AI assistants (Claude Desktop / Claude Code / any MCP client). Typed Records store, document knowledge graph with exact multi-hop traversal (ownership/relationship tracing), semantic + structural recall, deterministic corrections, provenance/citations, GDPR/Law-25 atomic forget, snapshot/replay audit. Thin client: all algorithms run server-side. For Neruva integration clients only -- NOT open source. Bring your own LLM.
pypi Proprietary -- (c) Clouthier Simulation Labs. For Neruva integration clients onl -
apudgithub.com/cadeos/apudactive
IIIF-grounded, citable vision for historical documents: region crops for VLMs, W3C annotations out. Python library + MCP server.
Apache-2.0 -
An MCP server that provides real-time information using Google Gemini's grounding capabilities
npm MIT ★ 8