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.
-
CLI and MCP adapter connecting acceptance criteria to JUnit evidence and validating structured implementation proposals.
MIT -
evidence-enrichment-enginegithub.com/tinnlo/evidence-enrichment-engineupdated this year
Replay-first evidence enrichment pipeline with hierarchical retrieval, schema-validated typed extraction, AI FinOps cost governance, execution policy capability control (off|audit|enforce), Langfuse-first observability, optional LangGraph retrieval, and MCP tooling.
-
Transcript-evidence prototype: YouTube identity normalization, conservative caption providers, offline supplied-evidence validation and explicit uncertainty.
-
Evidence-first CLI and optional MCP server for timestamped video transcript search and citation validation.
MIT ★ 1 -
timeline-truthgithub.com/hilmimuktitama/timeline-truthupdated this year
Evidence-preserving timeline compiler, strict validator, and schedule diff for AI-agent TPM workflows.
npm MIT -
Technical case study of a bounded MCP page inspector: FAST and rendered evidence, mediated networking, source attribution, OAuth and validation gates.
-
looktwice-mcpgithub.com/eason4kim-rocket/looktwice-mcpupdated this year
MCP server for the LookTwice API: email, EU VAT, IBAN, and domain validation plus evidence-backed CPSC recall screening from any MCP client.
npm MIT -
evidra-lockgithub.com/vitas/evidra-lockupdated this year
MCP Kill-switch for AI agents. Validates infrastructure operations before execution. Fail-closed. Evidence-backed.
Apache-2.0 ★ 8 -
Business research and validation for agents: authorized data collection, analysis, evidence audits and traceable deliverables. CLI, Skill and MCP.
MIT ★ 1 -
Check proposed agent actions against explicit policy and evidence with a local validator, GitHub Action, and MCP server.
npm Apache-2.0 ★ 1 -
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 -
opentargets-mcpgithub.com/openpharma-org/opentargets-mcpupdated this year
MCP server for Open Targets genetic evidence and target validation
NOASSERTION -
Local code validation with execution evidence, a CLI, and MCP tools.
npm MIT -
@trigvale/mcpgithub.com/sydacos-development/trigvaleupdated this year
MCP server for Trigvale — validate startup ideas before your build agent scaffolds code. One tool: validate_idea, returns 10-dimension scorecard + kill/pivot/test/build verdict + live evidence.
npm MIT -
medical-calc-mcpgithub.com/u9401066/medical-calc-mcpupdated this year
🏥 MCP Server with 121 validated medical calculators for AI agents. DDD architecture, evidence-based formulas with PMID citations. Supports Claude, GPT, and other LLM integrations.
Apache-2.0 ★ 5 -
Local-first Android reporter that turns bug reports into portable, reviewed evidence parcels for humans and coding agents. Offline validator, viewer and read-only MCP. AGPL-3.0-only; source candidate, not yet released.
Apache-2.0 -
organoid-mcpgithub.com/tan412/new_organoidupdated this year
MCP stdio shell for the organoid embodied-robot dataset QA kernel: 9-format lossless ingestion, 10 validators, six-state claim ledger. Perception-grade honesty: not_evaluated means insufficient evidence, never failure.
npm MIT -
Turn any website into validated, read-only tools and skills for AI agents (MCP). Evidence-first discovery, API-first tool generation, replay validation.
MIT -
resolvegithub.com/SamueleOlindo/resolveupdated this year
Evidence-first technical-support agent. Hybrid RAG (dense + BM25 + Reciprocal Rank Fusion), an agentic LangGraph workflow that self-corrects, and deterministic citation validation. Retrieval evaluation (Recall@k, MRR, nDCG), FastAPI, React, MCP server, Docker, Groq LLM.
-
MCP-powered marketing campaign review: deterministic checks plus an evidence-validated LLM reviewer. Synthetic data, demo only.
MIT -
Evidence-backed validation of repository memories, with reviewable corrections, a Python SDK, CLI, and read-only MCP server.
MIT -
Open-source, local-first AI scientific workbench for COMSOL, featuring multi-LLM support, hybrid RAG, optional MCP integration, intelligent modeling, controlled simulation, model readback, and evidence-based validation.SKY Workbench是一款开源、本地优先的 COMSOL AI 科学建模工作台,集成多大语言模型、混合 RAG 知识检索、可选 MCP 接入、智能建模规划、受控仿真执行、模型状态读回与基于证据的科学验证,致力于构建可靠、可追溯的 AI 辅助科学仿真工作流。
Apache-2.0 -
AI-first external attack surface assessment with continuous discovery, evidence-based vulnerability validation, and developer-agent MCP integration.
GPL-3.0 ★ 738 -
Local execution coordination for OpenAI Codex. Scoped, evidence-backed tasks based on technical findings with MCP integration and validation.
★ 1 -
Validate SOAR playbooks against BOTS v3 evidence using Splunk MCP, MITRE ATT&CK, VirusTotal, and approval-gated dry runs.
MIT -
SQL Sentinel · MCP-based MySQL query diagnosis for coding agents: scoped SQL, isolated shadow validation, graded evidence, and a web console.
GPL-3.0 ★ 4 -
agent-guardrailsgithub.com/logi-cmd/agent-guardrailsupdated this year
Checks AI-generated code changes before merge: scope, validation, risk, review evidence, and optional Verity receipts.
npm MIT ★ 28 -
Turn OpenAPI contracts into deterministic, stateful virtual services, synthetic test data, and validation evidence—with REST, MCP, and a local AI Copilot.
pypi MIT -
Evidence-based website scanner for what AI agents can read, answer, and call—validating MCP, WebMCP, APIs, and crawler access.
AGPL-3.0 -
cwgithub.com/Queopius/cwactive
Governed AI-assisted software development with Codex: phased execution, deterministic validation, independent reviews, and evidence-based approval gates.
Apache-2.0 ★ 2 -
A modular registry for trustworthy, evidence-based genomic analysis powered by Model Context Protocols (MCPs). Bridging AI and clinical genomics with traceable, validated data retrieval.
MIT ★ 1 -
🔌 Lets an MCP client check drug interactions, validate dosages, assess patient risk, and generate safety alerts. Uses evidence-based reference data. Not a substitute for clinical judgement.
npm MIT -
bullet-mcpgithub.com/nikkoxgonzales/bullet-mcpupdated this year
MCP server for evidence-based bullet point validation. Scores lists against cognitive research (Miller's Law, serial position effects) with actionable feedback.
MIT ★ 1 -
MCP server and agentic control layer for KiCad 10, combining LLM-based planning where useful with deterministic execution, state tracking, evidence, rollback, and KiCad validation.
AGPL-3.0 ★ 15 -
agentic-deep-auditgithub.com/th3vib3coder/agentic-deep-auditupdated this year
Evidence-first repository audit engine for local & GitHub projects. Python CLI with safety model, provenance tracking, supply-chain/license/risk analysis, MCP handoff, schema validation and portable audit artifacts.
-
Phionyx self-governance gate for Claude Code — three-layer verification (LLM declaration → git diff truth → deterministic gate) over the agent's own 'fixed / tested / changed' claims. v0.3.0 adds opt-in evidence binding: the require_tool directive (a factual claim without same-turn evidence must bind a tool first) and continuity binding (catches 'read-but-not-bound'). Both default-OFF / non-regressive; detector calibration is not yet validated.
pypi AGPL-3.0-or-later -
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.
-
Mechanically validated frontend aesthetics for AI agents: 89-palette evidence library, OKLCH generation, WCAG/APCA contrast, color-blind-safe data colors, AI-slop detection, theme packs, W3C design tokens. Skill + MCP server.
MIT -
MCP director for the forge pipeline: tells Claude which phase to run, validates each phase with real evidence, and persists state in a global SQLite DB to resume across sessions. Claude is the executor.
npm MIT -
mcp-sql-guardgithub.com/tahasiddiquii/mcp-sql-guardupdated this year
Governed read-only text-to-SQL MCP server: sqlglot AST validation (single read-only allow-listed SELECT), column-level PII masking, LIMIT and row caps, and a tamper-evident audit log. Offline DuckDB warehouse, governance-gated.
★ 1 -
MCP server for AI-assisted DFIR and incident response with Volatility, Eric Zimmerman tools, Sleuth Kit and PCAP analysis, featuring evidence integrity, audit trails, human validation and reproducible forensic workflows.
AGPL-3.0 ★ 1 -
Verifiable compliance evidence for agent decisions. MCP server for the OilFlow Network APIs: screen counterparties against 8 sanctions lists and a first-party fraud-cluster registry, check trade tradability across 235 jurisdictions, validate letters of cr
npm Apache-2.0 -
Agent-native source control: crash-safe integration, validation bound to the exact tree, policy-bound autonomy for AI agents, and a tamper-evident audit trail. Rust core, server, Web UI, MCP.
Apache-2.0 -
Author, generate, validate and govern AI agent rules, skills, plugins and MCP configuration across 52 coding assistants. Review with verified model evidence, evaluate skills, lock and sign dependencies, and publish bundles. CLI, MCP server and Go API.
npm MIT ★ 171 -
open-mind-mcp-servergithub.com/HelloThisWorld/open-mind-mcp-serverupdated this year
MCP server that exposes Open Mind's verification-first codebase context to AI agents — glossary, architecture, flows, and claim validation, with file:line evidence for every answer and honest refusals instead of hallucinations.
MIT ★ 1 -
PR-SENTINELgithub.com/Kapish7717/PR-SENTINELupdated this year
An autonomous GitHub App review agent built with LangGraph, FastMCP, and Groq (LLaMA 3.3). Combines deterministic static analysis (Ruff) with advanced LLM reasoning to catch bugs, styling nits, and security risks—validated with evidence-checking to prevent hallucinations. Runs asynchronously via webhook integration.
-
Anonymous MCP/HTTP: static Python artifact/source/consumer import-layout evidence; PyPI wheel metadata; mathematical IPv4/IPv6 set differences; deterministic probe planning. Offline ledger tools also available. Hosted inputs public/non-secret; not a resolver, compatibility check or routing/security validation.
MIT -
verified-unreal-mcpgithub.com/Pochomason/verified-unreal-mcpupdated this year
A verification layer for AI agents driving Unreal Engine 5.8: Claude has to PROVE a change works with evidence (fresh-process reads, validated screenshots, 3-source verdicts) instead of claiming 'done'. Built on the native UE MCP, for indie game devs.
MIT -
The local intern for Claude Code — 44 job-shaped tools, evidence-first briefs, durable artifacts. Local-first MCP server over Ollama with zero egress by default; no GPU big enough? Ollama Cloud runs the same 44 tools on 600B-class models, with automatic local fallback. Validated Hermes Agent integration on hermes3:8b.
npm MIT ★ 2 -
CAD2GIS · CAD to GIS for AI agentsno update data
Convert DWG/DXF drawings into reviewable GIS and QGIS delivery with source hashes and validation evidence. CAD2GIS 0.8.0 exposes 63 local MCP tools; four wrappers add 15 verified optional CAD tools for source-preserving 2D edits through an isolated MIT runtime. Supports Codex, Claude Code, Cursor and VS Code through repository plugins or stdio MCP configuration. Original code: Apache-2.0. Requires uv and a selected local project root; optional CAD runtime is installed separately. Demo cases remain CONDITIONAL; absolute surveying accuracy is not independently verified. 中文:把图纸转换为可以审查和交接的 GIS 成果,保留原图、记录修订,并将 CAD 编辑返回统一的 GIS 检查流程。安装与中英文 Wiki:https://mola-maker.github.io/CAD2GIS/wiki/
smithery -
SAP ABAP MCPgithub.com/Coaspe/sap-abap-mcpno update data
SAP ABAP MCP is a local-first Model Context Protocol server that enables AI coding agents to develop, test, analyze, and operate SAP ABAP systems through the official ABAP Development Tools HTTP services. It supports repository search and source editing, structured DDIC workflows, enhancement inspection, paged package navigation, source-backed object creation, semantic services, activation, ABAP Unit and ATC, transports, abapGit, RAP generation, debugging, bounded class and executable-program profiling, runtime dumps, traces and feeds, cross-system comparison, dependency analysis, guarded refactoring, and an opt-in same-origin Screen/GUI Status bridge. Curated compact, development, and assurance presets reduce tool-schema token use without removing capabilities from the default surface. The default local stdio mode can be extended with governed self-hosted Streamable HTTP, API-key roles, OIDC/JWT authentication, request-scoped SAP bearer forwarding, per-session isolation, rate limits, structured audit logs, or a side-effect-free embeddable Node.js API. It also includes browser OAuth Authorization Code with PKCE, a read-only transport assurance CLI and GitHub Action, BTP service-key profile import, and live-SAP evidence tooling. SAP profiles and credentials remain on the user's computer, with passwords and OAuth credentials protected by macOS Keychain or Windows DPAPI. SAP-dependent capabilities are validated against the user's own SAP release, configuration, and authorizations.
smithery -
MCP server for validating signed agent execution receipts.
pypi MIT -
Research first, build second: CLI and MCP tools for collecting primary sources, checking provenance, and handing traceable research to AI agents. Plain Node. Connected to MoonAliza and SmartRouter Repo.
NOASSERTION ★ 4 -
Deterministic, offline Chromium & Electron LevelDB (.log) carver, PNG/JPEG screenshot & ICC provenance inspector, ISO-13616/Luhn financial checksum validator, and zero-dependency Model Context Protocol (MCP) server.
pypi MIT -
@qualequest/orchestration-mcpgithub.com/qualequest/orchestration-mcpupdated this year
Model Context Protocol server for autonomous AI agent orchestration with 30+ coordination tools
npm MIT -
evidentiagithub.com/kgraph57/evidentiaupdated this year
Open-source medical citation verifier: catches fabricated DOI/PMID/NCT references with CLI, MCP, and Claude Code skill.
MIT ★ 1 -
agent-output-guard-mcpgithub.com/agenson-tools/agent-output-guard-mcpupdated this year
Revolutionary MCP server that solves the 36.9% coordination failure rate in multi-agent systems. Zero LLM costs, pure computation. Built on UC Berkeley MAST study evidence.
npm MIT -
NL→SQL agent over a lakehouse with five deterministic control rings: AST allowlist guard, cost budget estimated before execution, role-based masking and a tamper-evident audit chain. Served over MCP.
Apache-2.0