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.
-
llm-deliberate-mcpgithub.com/haza/llm-councilupdated this year
MCP server for multi-LLM deliberation with anonymized peer review
pypi MIT -
deliberate-aigithub.com/shanmukhaditya/deliberateupdated this year
Deep Ideation, Multi-Agent Deliberation & Systematic Thinking Engine for AI Coding Agents
npm MIT -
deliberate-thinkinggithub.com/kljensen/deliberate-thinkingupdated this year
An MCP tool for deliberate thinking (er...basically Rust clone of sequential-thinking)
★ 7 -
MCP server for structured multi-AI discussions — deliberate across Claude, GPT, Gemini and more before committing to decisions
npm MIT -
@iflow-mcp/haasonsaas-deliberate-reasoning-enginegithub.com/haasonsaas/deliberate-reasoning-engineupdated this year
MCP server that transforms linear AI reasoning into structured, auditable thought graphs
npm MIT -
Ask Codex, Gemini, Grok, and 400+ OpenRouter models (Qwen, Kimi, DeepSeek) for second opinions or arbiter-mediated consensus. One MCP server for Claude Code, Codex, Cursor, Kiro, OpenCode. Measures the models' usefulness, too.
npm MIT ★ 183 -
llm-council-mcpupdated this year
LLM-based council/consensus tool for multi-persona AI deliberation
pypi MIT -
VS Code extension for mumo — multi-model deliberation as MCP tools
MIT -
Cursor plugin for mumo — multi-model deliberation as MCP tools, with an auto-triggering skill
MIT -
jurygithub.com/anixes/juryactive
Resilient multi-model deliberation mesh, MCP stdio server, and zero-token code synthesis engine
MIT -
boardroom-bot-mcpgithub.com/chrisberno/boardroom.botupdated this year
MCP server and JS client for boardroom.bot — AI-powered multi-model deliberation
npm MIT -
Open-source LLM deliberation and decision-review engine with a Python CLI and MCP server. Includes minority reports and auditable transcripts.
NOASSERTION -
Give your AI a soul — open-source identity, memory & deliberation framework for AI employees. 40 MCP tools · 16 memory modules · 9 deliberation modes. 给你的 AI 赋予灵魂。
MIT ★ 2 -
Deliberately-vulnerable, fully mocked (network-less) reference MCP agent — the test target for the Mylonite AI-security scanner. NOT for production use.
pypi Apache-2.0 -
vechegithub.com/cyberash-dev/vecheupdated this year
MCP server for bounded multi-agent committee meetings — Codex + Claude Code deliberate in rounds and self-extinguish on <PASS/>
npm MIT -
Ask a panel of AI models, not one. Multi-model deliberation (debate, red team, forecast, consensus) as a personal MCP server, with the web app and shared engine.
Apache-2.0 -
@iflow-mcp/adoslabsproject-gif-nothumanallowedgithub.com/adoslabsproject-gif/nothumanallowedupdated this year
Epistemic dataset generation engine — 38 AI agents, multi-round Geth Consensus deliberation, 34 MCP tools, knowledge grounding (2.6M verified facts), zero-knowledge protocol, Ed25519 auth, SENTINEL WAF
npm MIT -
Deliberation and Decision Making MCP Server
MIT -
wisepanel-mcpgithub.com/ikoskela/wisepanel-mcpupdated this year
MCP server for Wisepanel multi-agent deliberation platform
npm MIT ★ 4 -
quorum-mcpgithub.com/aj-geddes/quorum-mcpupdated this year
Multi-provider AI consensus orchestration via MCP - Get better answers through deliberation
MIT ★ 3 -
vulnerable-notes-mcpgithub.com/bishnubista/vulnerable-notes-mcpupdated this year
Deliberately vulnerable MCP server for testing SAFE-MCP scanner - DO NOT USE IN PRODUCTION
-
A deliberately malicious MCP server for E2E testing purposes
npm MIT -
mcp-councilgithub.com/epicsagas/mcp-councilupdated this year
MCP server for multi-LLM peer review and council deliberation workflow
cargo Apache-2.0 -
symposiumgithub.com/AI-et-al/symposiumupdated this year
Multi-model deliberation as an MCP server — dissent surfaced verbatim, ledger in Markdown you own
AGPL-3.0 -
amogusgithub.com/rexheng/amogusupdated this year
A decision layer for AI agents — structured multi-perspective deliberation via MCP (Among Us themed)
MIT -
openrouter-fusion-agentgithub.com/chumikov/openrouter-fusion-agentupdated this year
Multi-model deliberation agent for OpenRouter Fusion on free models — CLI + MCP server for opencode
pypi MIT -
Loopback MCP for Enshrouded: inspect worlds, plan bounded admin builds, verify results and recover deliberately. Experimental enshwick mod.
MIT -
A deliberate way to expose registered WordPress abilities to the default MCP server. Free utility by Respira.
GPL-2.0 ★ 1 -
MCP server for IMAP mailboxes: read, search, organise and draft mail — it deliberately cannot send
npm MIT ★ 4 -
MCP-native LLM councils for debates, juries, blind panels, voting, refinement, and custom multi-model deliberation.
npm NOASSERTION ★ 1 -
MCP-Hardenedgithub.com/Nu1lstep/MCP-Hardenedupdated this year
A deliberately small MCP server demonstrating security hardening against the OWASP MCP Top 10, with a test suite that proves the defenses hold.
MIT -
A deliberately small GoHighLevel MCP server. Six tools, scoped to one job: knowing who you're about to talk to.
npm MIT -
llm-council-mcpgithub.com/JeremyGracey-AI/llm-council-mcpupdated this year
Multi-model LLM Council (Karpathy-style 3-stage deliberation over OpenRouter), exposed as an MCP server for Claude Code.
MIT ★ 1 -
sheets-mcpgithub.com/Abydin/sheets-mcpupdated this year
A deliberately small Google Sheets MCP server. Allowlisted spreadsheets, RAW-only writes so a cell cannot become a live formula, no Drive scope, stdio only.
MIT -
Opening the LLM Black Box: Kathara 12-Sphere Circuit Plugin for Nous Hermes with real-time visual route deliberation & MCP bridge
★ 1 -
Secure, privacy-focused MCP server for FactorialHR. Access employee directories & org structure in Claude - deliberately excludes payroll/financial data for enterprise security.
npm MIT ★ 3 -
conclavegithub.com/cogspro1234/conclaveupdated this year
An MCP server that lets Claude orchestrate Codex and Gemini via their CLIs — multi-model deliberation like a papal conclave.
GPL-3.0 -
A Codex plugin and local MCP server for deliberate context compaction in VS Code, with a verified task handoff and one-time automatic continuation.
MIT ★ 1 -
A deliberately vulnerable MCP server built as a synthetic test fixture — not for installation or connection. Ships the audit report mcpaudit returns for it.
MIT -
Multi-model deliberation as an MCP tool: several LLMs debate a question across rounds, then a chair synthesises consensus. Works with any OpenAI- or Anthropic-compatible provider.
★ 4 -
quorumgithub.com/marklynd/quorumupdated this year
A council of language models that scores a claim against a rubric, using only evidence you supply. Bounded timeouts, anonymised deliberation, hash-verified transcripts, MCP server.
MIT -
Council-Of-HALgithub.com/OmarTheGrey/Council-Of-HALupdated this year
An Open-Source AI Debate and Deliberation Engine offered in MCP, enabling consensus based advanced reasoning where AI models debate and refine positions across multiple rounds to reach the fullest most well-rounded possible response.
MIT ★ 1 -
MCP server for Bridle: lets an AI agent read its spending rules, check a payment, and propose one. Owner actions are deliberately not exposed.
npm MIT -
A Cloudflare Workers-native refactor inspired by @nautis/ghost-mcp-server, deliberately limited to secure Ghost post CRUD: list, read, style context, create, update, publish/schedule, and delete.
MIT -
sarvam-mcpgithub.com/harshil1502/sarvam-mcpupdated this year
Type-safe MCP server for the Sarvam AI API — chat, translation, transliteration, language detection, STT, and TTS across 10 Indic languages + English. 6 tools, deliberately scoped.
MIT -
A callable memory-and-divergence substrate for AI systems: an attributed multi-intelligence corpus, cross-model divergence atlas, deliberation engine, and open MCP/API
Apache-2.0 -
Caucus — a supervised hub where multiple AI agents (any MCP client) deliberate while a human operator watches live and can pause or stop the exchange.
pypi MIT ★ 1 -
Model Context Protocol server for the enterprise freelance marketplace. Points any MCP-capable client (Claude Desktop, Claude Code, Cursor) at the freelance subgraph with a platform API token registered as a freelance agent identity. Deliberately outside
npm MIT -
Aye Pulse market intelligence plus Aye Scout pre-payment diligence over x402: a free ecosystem snapshot and two deliberately scoped paid tools, with manifest trust verification and spend caps.
npm MIT -
MCP server exposing the eagle-bridge API as ten tools, so an agent can push, tag and print-process design assets without the bridge's bearer token ever entering a prompt. Deliberately thin — all the logic lives in eagle-bridge.
MIT -
amogusgithub.com/ViktorSmirnov71/amogusupdated this year
A decision layer for AI agents — structured multi-perspective deliberation via MCP. Among Us-themed council of 4 AI archetypes that debate, challenge & vote on plans before execution. Finalist — 2nd place @ Cursor Hack London 2026
MIT ★ 2 -
MCPGoatgithub.com/SabyasachiDhal/MCPGoatupdated this year
Deliberately vulnerable MCP server (aka MCP Goat) for security training — 26 challenges across 4 difficulty levels (incl. a secure reference), a victim-agent harness, and one-command Docker deploy. Practice penetration testing against the Model Context Protocol.
MIT ★ 12 -
Model Context Protocol server for Bitbucket Cloud pull requests - list, read, diff, comment, review, merge and decline - with OAuth 2.0 browser login and a deliberately small, auditable dependency tree. Native AOT binaries are published to GitHub Releases; this package is the `dnx bitbucket-mcp` convenience channel.
nuget MIT -
spring-ai-mcp-servergithub.com/SNagarjuna07/spring-ai-mcp-serverupdated this year
MCP server exposing task-management tools over the Model Context Protocol using Spring AI 2.0.0. Lets Claude Desktop call and manage real tasks in Postgres directly, with destructive delete deliberately kept outside the AI-reachable surface.
★ 1 -
Apple Ads MCP server + Python client. An AI agent can read the ad account and propose changes; every write goes through a preview a human approves, backed by guardrails, an append-only ledger and one-call revert. Built on the Apple Ads Platform API (v1), with a break-glass CLI for what the agent deliberately cannot reach.
MIT -
Model Context Protocol server for SonarQube Cloud - projects, quality gates, issues, security hotspots, measures and coverage, plus issue triage - with a deliberately small, auditable dependency tree. Native AOT binaries are published to GitHub Releases; this package is the `dnx mcp-sonarqube` convenience channel (the id differs from the product name because SonarSource holds both the SonarQube* and SonarCloud* reserved package-id prefixes on nuget.org; the installed command is still `sonarqube-mcp`).
nuget MIT -
mcpno update data
Read-only affiliate attribution and commission data for **iOS apps (StoreKit 2)** and **Shopify stores**. Attribloom attributes App Store subscriptions to the affiliate who drove them using the StoreKit 2 `appAccountToken` plus App Store Server Notifications v2, with no device fingerprinting. This server lets an agent read your campaigns, conversions, matured commissions, and payout eligibility. **Two of the seven tools need no API key**, so an agent can learn the integration before you have an account: | Tool | Needs a key | | --- | --- | | `get_started_guidance` | no | | `connect_store_guidance` | no | | `whoami` | yes | | `list_campaigns` | yes | | `campaign_readout` | yes | | `list_payouts` | yes | | `payout_eligibility` | yes | Money mutations (approving affiliates, creating campaigns, approving payouts) are deliberately not exposed. This server is read-only by design, and every figure it returns is read from the real ledger. Get an API key from the Attribloom dashboard under Settings, then API keys.
smithery -
MCP for AI agents that use Harmonica — create, query, and analyze your collective dialogues programmatically
npm MIT ★ 3 -
MCP server that lets an AI agent take part in Context Engine sessions as its own participant
npm MPL-2.0 -
@sylphx/consultant-mcpgithub.com/sylphxai/consultant-mcpupdated this year
Beta MCP server for typed SOTA consultation, model fan-out, and judge synthesis for autonomous agents.
npm MIT