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 descriptions, licences and links are read from each package's own registry (npm, PyPI, crates.io, NuGet, RubyGems; Go and Maven packages are only checked to exist), and packages the registry no longer has are removed. 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.
-
Vendor-agnostic, file-based governance & traceability layer for AI coding agents: decisions, shared memory, task state, learnings and token-aware context retrieval. Everything is persisted in English.
npm MIT -
Detect stale source context and retain traceable code-chain understanding for AI coding agents.
pypi -
slotledgeractive
Project-state MCP server for Claude Code / Codex: every AI decision visible, traceable, revocable.
npm -
scriptoriumgithub.com/iwizsophy/scriptoriumupdated this year
An MCP runtime for grounded Markdown and code retrieval with traceable references for AI-assisted development.
MIT -
@phmotad/fire-codegithub.com/phmotad/fire-codeupdated this year
Intelligent MCP execution engine for AI coding agents — persistent memory, context-aware, Git-traceable
npm MIT -
Project-state MCP server for Claude Code and Codex: every AI decision visible, traceable, revocable.
-
Grounded research for AI coding agents: Claude Code + Google NotebookLM over MCP, with traceable sources and a human in the loop.
MIT -
Agent-Integration-with-MCP-Servers-Notion-Git-github.com/Ashishsharmastudio/Agent-Integration-with-MCP-Servers-Notion-Git-updated this year
An Intent-to-Code Traceability Protocol utilizing MCP servers to force AI agents to align GitHub commits with Notion product specs.
★ 1 -
A local, traceable project memory for AI coding agents — MCP server + web UI over your Markdown, ADRs and agent memories
MIT ★ 1 -
Make AI research agents accountable — give every conclusion a traceable argument graph. MCP server + Claude Code plugin for paper reproduction, hypothesis verification, and auditable scientific reasoning.
MIT ★ 3 -
Spec-driven development for humans and coding agents. Requirements as plain files in Git, traceable from need to test, editable via desktop app, CLI, MCP and API. ReqIF import/export, baselines, variants.
NOASSERTION ★ 1 -
Open-source literature verification for Claude Code and Codex, with traceable source passages.
Apache-2.0 -
Inspectable, source-traceable public-equity research for Codex, Claude Code, OpenCode, Grok Build and ChatGPT Work developer mode. 26 provisional method lenses, 34 MCP tools. Research software, not investment advice.
npm MIT ★ 3 -
A local-first collaboration board for AI coding agents, with reliable cross-host handoffs, traceable evidence, and independent review.
MIT ★ 2 -
pyvisualizergithub.com/haider1998/pyvisualizerupdated this year
A deterministic, AST-verified call graph for any Python project — every edge traceable to a file:line. The reproducible, CI-gateable alternative to LLM-generated diagrams. No code executed. Nothing leaves your machine.
MIT ★ 12 -
Where every UI element lives in source — a static traceability map between iOS accessibility elements and Swift code. CLI + MCP server, no LLM, fully local.
MIT -
A robust MCP server that eliminates temporal blindness in AI coding agents through intelligent time tracking, persistent memory, and complete session traceability
MIT ★ 8 -
agentic-sdlc-mcp is an SDLC orchestration layer and control plane built on the Model Context Protocol (MCP). It wraps GitHub APIs into high-level, opinionated tools that enforce traceability, human-in-the-loop gates, quality thresholds, and security checks for AI coding agents.
npm MIT ★ 1 -
Cross-platform PRD verifier for Codex, Gemini CLI, Claude Code, CI, and other MCP hosts. Deterministic Hard Output Rules, cross-section traceability, and a full 9-file PRD pipeline with calibrated multi-judge consensus where the host supports orchestration. 17 MCP tools · 1,503 tests.
mcpb MIT ★ 7 -
For an organization to trust AI with its code, three things must hold — trust, traceability, and stability at scale. cladding wraps your AI coding agent: your intent goes in before it writes, and the result is verified against your spec after, so those three are earned, not assumed. First L4 implementation of the Ironclad standard.
MIT ★ 16 -
MCP server that maps cross-repository integrations, detects contract breakage, and visualizes it as a live graph — for AI coding agents (Claude Code, Cursor, Windsurf) and CI.
MIT ★ 1 -
Drop requirements docs → AI agents debate → structured specs with @implements in 7 languages.
Apache-2.0 ★ 6 -
agentwolfgithub.com/PercivalLin/agentwolfupdated this year
Agent-neutral MCP runtime for virtual engineering teams with workflow gates, global memory, and traceable ChangeSets.
npm MIT -
Stdio MCP server maintaining a JSON claims catalog for traceability, V&V, and FEVER auditing
Apache-2.0 -
Sign-off gate and requirement-to-test traceability for spec-driven AI development (OpenSpec + Superpowers compatible).
npm MIT -
Janus of Agents (formerly SpecGate): Sign-off gate and requirement-to-test traceability for spec-driven AI development (OpenSpec + Superpowers compatible).
npm MIT -
hrevn-mcp-servergithub.com/ai-human-andalusia/hrevn-mcp-serverupdated this year
Minimal MCP server for HREVN tools backed by the live managed API.
pypi MIT License -
Requirements, evidence and decisions for a codebase - computed, not declared. MCP server + CLI.
MIT ★ 1 -
Showcase of an AI copilot for Db2 incident investigations: it reads insurance data through a read-only MCP server, links every finding to evidence, and leaves the decision to the investigator. Demo video, screenshots and architecture decisions. Synthetic data; source code private.
-
Midasgithub.com/vornicx/Midasupdated this year
Local-first, eval-first memory for long-horizon AI agents — no LLM at ingest. Python SDK + MCP server with source-traceable recall, belief revision, selective forgetting, and reproducible benchmarks.
pypi MIT ★ 14 -
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