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.
-
CTXgithub.com/Alegau03/CTXupdated this year
CTX - Context Runtime Engine for Coding Agents
cargo ★ 146 -
ctx — a markdown context store for coding agents: validated writes, budgeted reads, audit trail. CLI, memory-tool handler and MCP server over one stdlib core.
MIT -
Context compression for AI agents — cut token usage costs with retrievable CCR compression. Rust core + Python API, Claude Code & Codex plugin, MCP server. PyPI: furl-ctx
Apache-2.0 ★ 5 -
Local-first code graph + context engine for AI coding agents (Rust/MCP server). Published in microsoft/winget-pkgs (winget install halloffame12.CTX) & awesome-mcp-servers. A+ on Glama.
npm MIT -
ContextDoc tools: run and generate conceptual tests (ctx-run), detect .ctx drift (ctx-watch), lint .ctx files (ctx-lint), expose them to agents over MCP (ctx-mcp)
pypi MIT -
DSH 的 Tandio 实验平台接入 bundle:expctl mcp(stdio)+ expd /mcp(Streamable HTTP),浏览器 device 登录换 read+draft Agent token(存 ctx.credentials),工具白名单收敛到课程/实验/待上线链路,prod 写操作走 ctx.approval;CABT-293 起带右侧课程编辑面板(Markdown 预览/受控编辑/版本对比,client 半边走 shell.overlay,浏览器不接触 token);C
npm UNLICENSED -
drop-mcpupdated this year
MCP server for drop.ctx.st — AI agents read & write E2E encrypted rooms
npm MIT -
ctxlayer-releasegithub.com/abhilashsblai/ctxlayer-releaseupdated this year
CTX Layer: context engineering, memory, governance, MCP, and impact analysis for AI coding agents like Codex.
NOASSERTION -
ctx — a queryable code graph for coding agents, across Rust, Python, TypeScript, Go and Java. Call graphs, blast radius, breaking-change gates, port parity. Every answer reports its own completeness. CLI + MCP server, one binary, no LSP, no index.
cargo MIT ★ 3 -
perseus-ctxgithub.com/perseus-computing-llc/perseusupdated this year
Live workspace context engine for AI assistants — resolve before context.
pypi MIT -
ctxgithub.com/kenmham/ctxupdated this year
Context layer for coding agents: open-source, free, extremely fast.
-
ctx-capturegithub.com/sathviknayak123/ctx-captureupdated this year
MCP server for context-window observability: capture and query exactly what an LLM agent saw at every step.
pypi MIT -
ctx-memorygithub.com/ghadisaab/llm-memoryupdated this year
Persistent memory layer for LLM coding agents (Claude Code, Codex, Gemini, OpenCode) — zero cloud, local SQLite + MCP
npm MIT -
Tool-neutral codebase context kit for AI coding agents (frontier to local LLMs): rules sync, ranked repo maps, profile-aware context packs, MCP server
npm MIT -
@ctx.com/cards402github.com/ctx-com/cards402updated this year
Virtual Visa cards for AI agents — pay USDC or XLM on Stellar, get a card in ~60s
npm MIT -
LeanCTX is the Context Gateway for AI Systems. Select task-relevant context, apply supported controls, connect local knowledge and inspect recorded evidence. Open-source Engine: CLI, MCP, hooks and supported proxy paths.
npm Apache-2.0 ★ 3,952 -
better-ctxgithub.com/jadzeino/betterctx-clientupdated this year
Context Intelligence Engine with CCP. 34 MCP tools, 8 read modes, 90+ compression patterns, cross-session memory (CCP), persistent AI knowledge with temporal facts + contradiction detection, multi-agent context sharing + diaries, LITM-aware positioning, AAAK compact format, adaptive compression with Thompson Sampling bandits. Supports 24 AI tools. Reduces LLM token consumption by up to 99%.
cargo MIT -
Shared handwriting whiteboard for humans and agents: an authenticated MCP gateway (OAuth 2.1, GitHub device flow) over the unmodified mcp-excalidraw-server.
MIT ★ 1 -
AI-first vertical (9:16) video editor engine — the LLM plans a declarative Timeline IR; a deterministic FFmpeg renderer executes. Drivable from Claude Code via MCP.
MIT ★ 2 -
TiMEM 长期记忆的 DeepSeek Harness 插件:MCP 桥(凭证走 ctx.credentials)+ 内置 5 个记忆 skill
MIT