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. Each candidate is checked by a language model from its name, description and topics, and only actual MCP servers are listed (not clients, SDKs, lists or skills). Package metadata comes from ecosyste.ms (CC BY-SA 4.0). 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.
-
incur-based CLI/MCP server exposing TypeSafe jev (systemOne) typed-question calls to agents
npm MIT -
Grade essays and resumes, and judge texts with typed yes/no, choice and score questions.
npm MIT -
knowuswell-mcpgithub.com/ppop123/knowuswell-mcpupdated this year
MCP server for the KnowUsWell friendship quiz — AI assistants host the two-player quiz: 16 behaviour questions, 16 pair types.
npm MIT -
MCP server exposing one tool: ask Jev a typed question and get a short answer back.
MIT -
mcp-server-jevactive
MCP server for Jev, TypeSafe's System One model: typed choice, score, and noul questions over MCP
npm MIT -
jev — the command-line interface and MCP server for the TypeSafe API. Ask questions about text and get typed answers with probabilities.
cargo MIT -
google-forms-mcp-extendedgithub.com/xari-ya/google-forms-mcp-extendedupdated this year
A secure, feature-rich Python MCP server for Google Forms. Create and manage forms, advanced question types, quizzes, media, publishing, and responses through 22 typed tools with least-privilege OAuth authentication.
MIT -
Local decision workflows for AI assistants: typed questions, scoped confidence checks, caching and MCP, powered by Ollama and Tev1.
MIT -
moodle-mcp-bridgegithub.com/nmaaalhawary/mcp-moodleupdated this year
MCP server exposing a Moodle site as 88 typed AI tools: read courses, grades and deadlines; build courses, quizzes, questions and content.
pypi MIT License -
MCP server + CLI: send files and typed questions to TypeSafe's Jev decision model, get the raw answer back. One session folder per request.
npm MIT ★ 1 -
Eval-first MCP server for TypeSafe's Jev: prototype typed questions, map them over many items, and measure them against labeled fixtures.
npm MIT -
MCP server for Jev (TypeSafe System One): the three official question types — choice, score, noul — plus batch classify. Calibrated probabilities, ~0.5s, <$0.001/call.
MIT ★ 3 -
A Jev-first memory layer for AI agents — frozen typed questions, deterministic policy, a sealed judgment log. Model proposes, the ledger decides.
MIT ★ 2 -
LangGraph ReAct agent that answers analytical questions about a customer support dataset via typed tools, with persistent memory and an MCP server
-
One rubric, every decision model. Run the same typed questions on Jev, Clef and OpenAI's Decisions API, and see where they disagree on your own cases.
npm MIT -
Go semantic tools for the MCP server built into GoLand: 24 tools that answer questions about Go code through the type system, not text search
Apache-2.0 ★ 4 -
An abstraction for System One models, currently supporting TypeSafe's Jev and Laya. Typed questions, probabilities, and "don't know" instead of a guess, from Python, the command line, or MCP
MIT ★ 1 -
mcp-surveygithub.com/zavora-ai/mcp-surveyupdated this year
Survey MCP Server — a survey platform (authoring with multiple question types & branching, distribution via collectors and invitations, response collection, and measurement/analytics including completion rates, NPS, CSAT, score distributions, and cross-tabs) with a full audit trail
cargo Apache-2.0 -
A Jev-compatible decision layer as an MCP server: typed noul/choice/score questions with probability distributions and confidence. Pluggable backends (local needle/laya models or any OpenAI-compatible endpoint), calibration tracking, conditions v1, eval harness.
MIT -
cevgithub.com/sshworld/cevactive
A decision tool for coding agents that would rather ask a human than guess. Typed questions in, a confident answer or an escalation out. Claude Code plugin + MCP, no API key.
MIT -
claude-cartographygithub.com/shumatsumonobu/claude-cartographyupdated this year
Give Claude Code a map, not a guess — code and docs as one knowledge graph, with 4 analysis tools auto-routed by question type. Docs in Japanese.
MIT ★ 1 -
Discuss backend architecture in pictures with your coding agent — picks the diagram type and zoom level for the question, then draws it on a live Excalidraw canvas. Agent skill, MCP server, and CLI.
npm MIT -
An MCP server that answers network troubleshooting questions ("why can't A reach B?") with deterministic, pre-shaped KQL against Azure Log Analytics — the agent picks typed tools and fills parameters, it never writes KQL itself.
NOASSERTION -
Read-only MCP server for hospital price questions: six typed tools over clear-pricer's pinned, hash-verified release of three Chicago hospitals' price files and the NPPES registry, every row cited to the hospital's own file.
npm MIT -
lextiva-mcpgithub.com/arturayupov/lextiva-mcpupdated this year
MCP server exposing the Lextiva compliance reference dataset — 18 privacy + AI regulations, 21 jurisdictions, 7 document types, 8 data-subject rights. Lets Claude/Cursor/any MCP-aware agent answer compliance questions with primary-sourced, verifiable data.
NOASSERTION -
Tiehall lets one team's coding agent hand a typed request to another team's agent, and carries the questions, status and commitments between them until a pull request closes the request, on a server your company runs.
AGPL-3.0 -
MCP server that lets Claude Code, Codex & AI agents drive a real Chrome browser with JEV (TypeSafe System One): ~100 ms typed decisions, calibrated confidence, human-in-the-loop questions, safety gates and a debug trace.
MIT ★ 7 -
Knowledge engine for AI agents: your documents become a git-versioned markdown knowledge graph the agent navigates over MCP. Curated summaries, typed edges, cited answers, SQL over your spreadsheets, no chunking. Same 12B local model: 0/11 on multi-hop questions as top-k RAG, 11/11 as a navigator. Python, Apache-2.0, self-hostable.
Apache-2.0 ★ 13