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.
-
—
mcpb MIT ★ 1 -
Local Jev-compatible decision model as a Claude Code assistant — Bash risk gate, Stop evidence gate, decide MCP tool, measured eval harness. Zero tokens, ~0.75 s per judgment, never the decider.
MIT -
preflightoraclegithub.com/ToolOracle/preflightoracleupdated this year
Decision Preflight MCP — Pay before you act wrong. 6 tools for AI agent decision gates. Returns go/caution/stop/insufficient_evidence in one call. Runs evidence, freshness, provenance, policy & red-flag checks in parallel. x402 pay-per-call via USDC on Base.
NOASSERTION -
shakerscan-mcpgithub.com/shakerscan/shakerscan-mcpupdated this year
MCP server for ShakerScan AI Gate, signed deploy decisions, and DAST workflows
npm MIT -
Human approval requests for agent actions: expiring, single-use decisions. Trusted executors must enforce the gate; not a same-user security boundary.
npm MIT -
MCP server & CI gate for release decisions: UI, API contract, API functional, LLM red-team and JUnit evidence under a written policy → ship, conditions, needs a person, or block, with an audit trail
pypi MIT -
trust-gate-mcpgithub.com/Cyber-Warrior-Network/trust-gate-mcpupdated this year
Policy-gated AI decisions with NIST FIPS 204 post-quantum receipts. EU AI Act Article 50 native. Every MCP tool call signed, auditable, provable offline.
Apache-2.0 -
intent-gate-servicegithub.com/baixinghao/intent-gate-serviceupdated this year
DingTalk sister MCP service of intent-gate: group consensus channel + blocking human-decision gates for requirement intent alignment. Optional — the main plugin closes the loop alone; this posts the questions to the people who should answer them.
MIT ★ 3 -
Stop AI coding agents from guessing. MCP server + Claude Code plugin that enforces intent alignment BEFORE coding: PRD → intent-confidence gate → Mermaid contracts (state machines / sequence diagrams / decision tables) → mechanical lint with zero CRITICAL → then code. Human-in-the-loop, file-persisted, zero-config.
MIT ★ 3 -
decreegithub.com/doruksahin/decreeupdated this year
Decision governance for code and the agents editing it — track PRD→ADR→SPEC decisions, map them to files, and gate changes against them. Deterministic, honest CLI + MCP.
MIT ★ 2 -
Phocinae MCP decision server: gate / classify / route / score tools backed by the local Phocinae model
pypi Apache-2.0 -
MCP server that returns calibrated probabilities for an agent's own decisions, with an act/escalate gate, across local Laya and remote Jev engines.
NOASSERTION -
kontrol-freek-mcpgithub.com/berkbayri/kontrol-freek-mcpupdated this year
AI assumption firewall for MCP agents — intercepts risky decisions, checks past decisions, detects contradictions, and gates human approval via Telegram or Slack.
pypi ★ 1 -
(Contextual Action & Verified Decision Autonomous eXecution) ————————————————— A deterministic FalkorDB graph enforcement layer and enclosed MCP gate that prevents AI coding agents from executing actions on stale, superseded, or unprovenanced architecture decisions.
npm NOASSERTION -
io.github.IndirectTek/mcp-serverno update data
Elixium AI — XP-native agile platform. TDD-gated stories, team decisions, learning loop.
npm -
coregithub.com/Smeldr/coreactive
Keeps decisions coherent when people and AI agents move fast. A searchable graph of your decisions: who decided, what is in force, what builds on what. Rights set per domain and area, MCP-native, optional merge gates. Self-hosted and open source, with nothing to install beside it.
go AGPL-3.0 ★ 10 -
Universal Model Context Protocol (MCP) server and CLI runner for TypeSafe Jev decision models. Calibrated evaluation for code reviews, acceptance gating, and agent guardrails across Claude Code, OMP, and Cursor.
MIT ★ 2 -
Verax: the accountable agent body. Gate, decision record, receipt, reconciliation. Apache-2.0.
npm Apache-2.0 -
brainsgithub.com/abhishek42638/brainsupdated this year
Decision OS for B2B service companies — an agent gathers evidence, deterministic rules and an approval gate decide, and every decision ships a full audit trail. MCP · hand-rolled agent loop · pgvector RAG · FastAPI · Cloud Run.
AGPL-3.0 -
TypeSafe AI Jev MCP server for bounded probabilistic decisions: four read-only stdio tools for choices, routing, risk signals, and yes/no gates. Conservative second opinions with installers for Codex, Claude Code, Cursor, Kimi, Gemini CLI, and other MCP clients.
MIT ★ 1 -
Pre-execution decision verification for AI agents. Sentinel → adversarial RV panel (SERV Reasoning) — ALLOW / BLOCK / UNCERTAIN with signed attestation. MCP server, x402-gated on Base.
npm MIT ★ 1 -
Effect gate for AI agents. An agent asks before it charges a card, ships a deploy, or sends the email, and gets back a durable decision — so the same real-world action is attempted at most once. Ratchet never performs the effect itself and holds no vendor credentials.
Apache-2.0 -
Design system governance over MCP. Thirteen tools, not a hundred: adoption evidence, blast radius, decision records, drift and CI gating.
npm MIT ★ 9 -
mcp-fraudgithub.com/zavora-ai/mcp-fraudupdated this year
Fraud MCP Server — unified fraud detection (signals + risk scoring) and case review (alerts, gated decisions, block/allow lists) for retail and insurance (SIU) fraud agents
cargo Apache-2.0 -
System One decision runtime for DeepSeek Harness: an MCP server and a DSH host plugin with automatic pruning, stagnation breaking and completion gates
-
Task backlog and notes board for coding agents on Herdr - claims with leases, evidence-backed review, and a human decision gate, in one Go binary.
MIT ★ 2 -
MCP server: control any Android phone over ADB (USB/wireless) with JEV, TypeSafe's fast decision model — typed UI understanding, explicit user consent, safety gates
MIT -
Page truth for browser agents — and decisions that replay, test and audit. Jev picks the element, risky actions are gated, every run replays offline in CI.
pypi Apache-2.0 ★ 2 -
@memnox/mcpgithub.com/memnox/mcpupdated this year
Memnox MCP server — guardrails for AI coding agents: validates plans against team decisions, blocks bad writes via hooks, and gates CI, backed by organizational memory
npm MIT -
MCP server for Jev (TypeSafe AI's System One model) — give any agent typed, calibrated decisions: classify, score, check, gate risky tool calls. Try free: jevtypesafeai.com
MIT ★ 3 -
Vetono update data
93 agentic MCP tools + 49 specialist agents for every major AI CLI — self-learning, no API keys. Veto gives Claude Code, Codex CLI, Gemini CLI, Cursor, Windsurf, Zed, and JetBrains a council of specialist agents, persistent cross-platform memory, a self-learning router, CI/CD gates, and dependency-hallucination + decision-drift guards — all powered by your existing AI subscription. No separate API keys, no extra billing. **Highlights** - **Council debate** — multi-agent GREEN/YELLOW/RED verdicts on risky changes (LLM-backed via MCP Sampling, with deterministic fallback) - **Cross-platform memory & session handoff** — save on Claude, resume on Gemini or Codex; nothing needs re-explaining - **Self-learning router** — improves task routing from recorded outcomes - **Guards & gates** — CI gate, diff review, secrets scan, dependency-hallucination verification, decision-drift enforcement - **Compact mode** — advertise 9 tools instead of 93 and save ~16K context tokens; every tool stays callable **Links**: [GitHub](https://github.com/jigyasudham/veto) · [npm](https://www.npmjs.com/package/@jigyasudham/veto)
smithery -
React & React Native decisions with receipts — a continuously verified corpus with a doctor CLI, migration planner, CI review gate, and MCP server for coding agents.
npm Apache-2.0 -
agentic-earnings-tradinggithub.com/vuolo/agentic-earnings-tradingupdated this year
Claude-orchestrated earnings-event trading: agent analysts over the Robinhood MCP, a server-side risk-gated paper broker, versioned trading policy, and a decision/outcome dataset feeding an eventual ML sidecar. AI/data-center universe first.
-
Reliable agent tool calling on small models: schema-validated decisions, tiered routing with evidence-based escalation, an MCP server and a CI eval gate. Python stdlib only.
-
scheduleroraclegithub.com/ToolOracle/scheduleroracleupdated this year
SchedulerOracle — Persistent Autonomous Task Scheduler MCP for AI Agents. 9 tools: cron/interval/one-time scheduling, auto-store results in MemoryOracle, optional Decision Preflight gate, execution history. The motor that makes agents work 24/7.
NOASSERTION -
Local-first continuity memory for AI coding agents. An MCP server that remembers decisions, open loops & findings across sessions. No cloud, evidence-gated, Apache-2.0.
Apache-2.0 -
Universal MCP server port of VibeWise: design-first gate for any coding agent (Claude Code, Cursor, Codex, Antigravity). Ask the user's approach, record decisions in .vibe-wise/, write code only after approval.
NOASSERTION ★ 1 -
mcp-confidencegithub.com/shaxzodbek-uzb/mcp-confidenceupdated this year
A drop-in confidence gate for LLM agents: turn token logprobs into an accept / verify / ask-a-human routing decision, with honest per-model calibration. Local & open-model first, plus an MCP server.
MIT -
buzz-axirugithub.com/axiruai/buzz-axiruupdated this year
Bounded spend authority for Buzz agents. A gating MCP proxy in front of your payment MCP server: policy evaluation before money moves, human approvals in the Buzz channel, and a hash-chained decision log keyed to the agent's Nostr pubkey.
npm Apache-2.0 -
mcp-credit-bureaugithub.com/zavora-ai/mcp-credit-bureauupdated this year
Credit Bureau MCP Server — a credit-reporting platform (consumers, credit reports with tradelines/inquiries/public records, scores and factors, permissible-purpose-gated report pulls that log a hard inquiry, FCRA disputes, monitoring alerts, and affordability/decision analytics) with FCRA-style permissible-purpose enforcement and full access logging
cargo Apache-2.0 -
cortexgithub.com/ProductionLineHQ/cortexupdated this year
Persistent memory for Claude Code. MCP server that captures decisions, context, and preferences during sessions and injects them back automatically. Local-first SQLite, quality-gated, multi-machine sync. Your AI starts every session fully informed.
npm MIT ★ 1 -
A Swiss Army knife that lets AI agents get the most out of decision models, simply. Typed yes/no, choice and score judgments over MCP, REST and a CLI, to gate and triage agent work. Written in Rust.
MIT ★ 2 -
Multi-tenant agent for telecom billing disputes: classifies disputes, retrieves tariff and contract clauses alongside invoice and call records, and makes cited credit decisions, with deterministic credit calculation, fraud escalation and CI-gated evals. LangGraph, MCP, fine-tuned classifier vs LLM baseline
MIT -
Local-first RAG engine built on the Laya decision engine: hybrid BM25 + dense recall, calibrated relevance re-ranking and answerability gating that return evidence with probabilities rather than generated prose. CLI, FastAPI HTTP API, MCP server and web GUI.
Apache-2.0 -
MCP server mang quy trình làm việc theo wave giữa PO và AI vào mọi dự án: Orchestrator điều phối, bộ phận lập theo dự án, decision log, quality gate, nâng cấp bằng three-way merge.
pypi MIT ★ 1 -
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 -
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.
npm MIT ★ 6 -
🔌 Exposes Per-agent-pair IAM for A2A. Define policies ('orchestrator may call billing only when amount<1000'), gate every A2A call via evaluate_call. EU AI Act Art 14 + ISO 42001 Annex A.7 evidence with signed policy-decision attestations.
pypi MIT -
Verified memory for coding agents: your rules, decisions and past defects, briefed before each task, proven by a check you choose, and re-opened when the code they cover changes. Prompts advise; the gate enforces. Claude Code hooks, MCP for any agent, one SQLite file in the repo.
pypi MIT ★ 4 -
concordance-enginegithub.com/matharrismma/lighthouseupdated this year
Four-Gates packet validation engine with cross-disciplinary computational verifiers and MCP server for AI assistant integration.
pypi MIT -
Servidor MCP para rastreamento de decisões arquiteturais com validação de predições e outcome gates.
pypi MIT -
MCP server that compiles checkable task contracts, retrieves quote-grounded web evidence, and verifies drafts for coding agents — especially cheap/small host models.
pypi MIT -
MCP server for TypeSafe AI's Jev: typed yes/no, choice and score answers with act / review / abstain gating.
npm MIT ★ 2 -
Multi-source clinical lab analysis pipeline: lab tests + PubMed evidence + MRI cross-validation, with DSPy dual-track LLM, quant gates and MCP server.
MIT ★ 1 -
nocapgithub.com/xi029/nocapactive
No evidence, no answer. 🧢 System One evidence gate for RAG & AI agents: Jev / Laya / any LLM judges before your LLM speaks. MCP server, Python SDK, CI route tests.
Apache-2.0 ★ 1