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.
-
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 -
Decision memory layer for AI-assisted codebases — trả lời 'logic nghiệp vụ này có từ bao giờ, ai quyết, vì sao'. Markdown là nguồn sự thật, SQLite FTS5 để search, MCP server cho AI agent.
MIT ★ 1 -
A proactive, headless contextual intelligence layer for the hospitality sector. It predicts the impact of a property's environment on guests' experience, optimizing decision making and operations.
MIT ★ 3 -
MCP server for PocketLantern — blocker-aware decision layer for AI coding agents, grounded in source-linked, time-sensitive facts
npm MIT -
Bridges the "black box" problem by creating a formal, inspectable decision layer. Each decision traces to explicit logical operations - no hidden states. Leverage NN perception while maintaining human-interpretable decision logic.
MIT ★ 2 -
gramatrgithub.com/gramatr/gramatrupdated this year
grāmatr — context engineering layer for AI coding agents. Every prompt gets a pre-computed intelligence packet: decision routing, capability audit, behavioral directives, memory pre-load, and ISC scaffolds. Continuity across sessions for Claude Code, Code
npm SEE LICENSE IN LICENSE -
Jev MCP server: a decision layer for Claude Code, Codex, and Cursor built on TypeSafe Jev (System One model). Ship gates, risk checks, file triage that keeps files out of your agent's context, and a safe headless browser.
npm MIT -
Decision-layer Whoop MCP for agents: intents, recovery-in-band checks, and training calls
-
probbit: a personality, mood and memory layer for AI agents, on an exact p-bit decision engine. Every stance and plan is computed on a CPU, keeps its rules by construction, and is replayable and verifiable. CLI, MCP server, OpenClaw plugin, npm, Python, wasm. Zero-dependency Rust.
npm Apache-2.0 ★ 1 -
pocketlanterngithub.com/pocketlantern/pocketlanternupdated this year
Blocker-aware decision layer for AI coding agents, grounded in source-linked, time-sensitive facts.
MIT ★ 4 -
HeartFlow (心虫) — AGI Layer 1: the Discriminator. Rule-based engine that checks what AI says before it reaches humans. 46 discrimination dimensions, 9-layer pipeline, 137 modules (logic/decision/cognition/emotion/memory/ethics/collaboration), 179 MCP tools
npm MIT -
@isaacriehm/cairn-coregithub.com/isaacriehm/cairnupdated this year
Cairn core — state + context layer. Curated `.cairn/ground/` (decisions, §INV invariants, canonical-map, brand, quality-grades), MCP server, init wizard, hook runners, sensors, GC drift sweep.
npm MIT -
Git-native context layer for AI coding agents. Your team's real engineering decisions — patterns, pitfalls, conventions — live as reviewed markdown files in your repo; agents consult them before writing code and record what they learn. Files-first, no server required; MCP as an optional serving layer.
MIT ★ 27 -
agent-safety-mcpgithub.com/LuciferForge/agent-safety-mcpupdated this year
MCP server wrapping ai-cost-guard, ai-injection-guard, and ai-decision-tracer into a single safety layer.
MIT ★ 1 -
Intentgithub.com/SixByFive/Intentupdated this year
Git-native reasoning layer for software projects. Track the why behind code with plans, decisions, constraints, and AI-readable project context.
MIT ★ 4 -
@contorium/mcpgithub.com/contoriumlabs/contoriumupdated this year
MCP bridge to Contorium Project Intelligence Layer — inspect, capture, and transfer project decisions, reasoning, and state for Claude Code, Cursor, Codex, and other MCP hosts.
npm MIT -
Causal memory layer for AI agents — MCP server that records decision→outcome relationships. Survives compaction.
Apache-2.0 ★ 107 -
@iflow-mcp/moonx010-hive-memorygithub.com/moonx010/hive-memoryupdated this year
Cross-project memory layer for AI coding agents — MCP server that maintains context, decisions, and knowledge across workspaces
npm MIT -
contoriumgithub.com/ContoriumLabs/contoriumupdated this year
Project Intelligence Layer for AI-assisted development. Preserve project decisions, reasoning, constraints, history, and evolving state across AI tools.
MIT ★ 7 -
The decision layer for personal AI. Local-first runtime where agents automate across GitHub, Slack, Gmail & 45+ services — but every risky action can stop for your yes, workers (agents) earn autonomy from track record, and every decision leaves a receipt. Your models, your machine, your policy. Includes an MCP bridge: 180 IDE tools for Claude Code.
npm MIT ★ 33 -
A fast decision layer for coding agents: Jev ranks files, picks tests, sorts failures and reviews diffs, so the expensive model reads only what matters. CLI, MCP server and Claude Code plugin.
MIT ★ 1 -
Paradigm is a procedural learning and trust layer for AI agents. It learns from validated deliberative decisions, compiles stable behaviors into specialized reflexes, and keeps uncertain or unsupported states on the deliberative path.
MIT ★ 2 -
toon-memorygithub.com/luiggival08/toon-memoryupdated this year
The Continuity Layer for AI Agents — an MCP memory server that preserves project knowledge, decisions, and conventions across sessions so every session starts where the last one ended. Offline, token-efficient, no LLM, no server; works with 20+ agents.
npm MIT -
⚓️ A shared knowledge layer for coding agents. Serves each agent only the decisions and conventions that govern the file it has open, over MCP — then records whether the code still honors them. Your source never leaves your machine.
AGPL-3.0 -
The open routing layer for AI tools: an MCP server that routes agent tool calls (search, extract, browse) across providers per policy — cost, latency, pinning, deny lists — with failover and the routing decision on every result. BYOK.
npm BUSL-1.1 -
Mynd is a universal memory layer for AI that automatically captures your digital context and streams it securely to any AI via Model Context Protocol (MCP). Your AIs finally remember everything about you - your decisions, preferences, history, and patterns - while your data never leaves your device.
★ 2 -
Jev MCP server: a decision layer for coding agents, built on TypeSafe Jev (System One model). Ship gates, risk checks, file triage that keeps files out of context, and a safe headless browser, with calibrated confidence. For Claude Code, Codex, Cursor.
MIT ★ 6 -
A personality, mood and memory layer for AI agents on an exact p-bit decision engine: an AI persona outside the model, every stance computed on a CPU, rules kept by construction, replayable and verifiable. Installs the prebuilt CLI and MCP server (probbit
npm Apache-2.0 -
The open routing layer for AI tools. MCP server that routes agent tool calls (search, extract, browse) across 8 providers per policy — cost, latency, pinning, deny lists — with failover and the routing decision on every result. One Go binary, BYOK, no account. Source-available, BUSL 1.1 → Apache 2.0.
npm NOASSERTION ★ 2 -
figma-edit-mcpno update data
Connect AI assistants to Figma via Model Context Protocol to read designs, create and modify elements, and manage design systems programmatically. This plugin empowers your AI assistant to become an active design co-pilot, executing updates and modifications Safer, Cleaner, and Faster than manual entry. By offloading repetitive, error-prone manual operations to automated systems, designers are freed to focus entirely on creative decision-making and product strategy. ### Core Principles * **🛡️ Safer**: The server enforces robust programmatic checks and validations that exceed the Figma Desktop app's default protections. For instance, it prevents deleting variables currently in use to avoid dangling references. By validating all writes against a locked editable scope and strict name checks before execution, it protects files from human error and AI hallucinations. * **✨ Cleaner**: System-driven, programmatic operations guarantee that no layers, text nodes, or components are skipped or forgotten during complex updates. Every design file stays consistent and conforms exactly to your design system. * **⚡ Faster**: Bulk edits like propagating instance overrides, batch-updating copy, or restructuring auto-layout frames are handled via high-performance APIs. Hours of tedious, click-heavy manual work are completed in seconds. ### How it Works (Prerequisites) This server runs locally on your machine. To operate, it requires: 1. The Figma Desktop App to be open. 2. The `Figma Edit` plugin to be active on the file you want to modify (listening on local port `3845`).
smithery -
Memolok is the Decision Record platform. It records decisions as Memolok Decision Records: structured, immutable entries capturing the need, the alternatives, and the reasoning behind a choice. Plugin for Claude and Claude Code.
★ 1 -
Jev (TypeSafe System One) sebagai partner keputusan untuk AI agent - agent skill lintas SDK + MCP server + implementasi terukur. Dokumentasi Bahasa Indonesia.
MIT -
open-indexgithub.com/DrDroidLab/open-indexupdated this year
Deterministic memory layer for your agents
MIT ★ 114 -
Your portable AI memory — preferences, decisions, and corrections that follow you across models, tools, and devices. Zero-dependency, local-first, 85% PrefEval accuracy.
MIT ★ 9 -
coregithub.com/medmcp-dev/coreupdated this year
Medical reasoning layer for AI agents — deterministic symptom risk signals, drug interactions, labs, ICD-11; HTTP API + MCP (stdio). Not a substitute for clinical care.
-
watercoolergithub.com/mostlyharmless-ai/watercoolerupdated this year
Watercooler: your team's shared reasoning layer for agentic coding. Preserve the "why" around your code.
Apache-2.0 ★ 14