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.
-
A Python Module Which Compacts LLM Prompts
pypi MIT -
sts2-compact-mcpgithub.com/chiang881/sts2-compact-mcpupdated this year
Token-efficient MCP proxy for Slay the Spire 2 via Gennadiyev/STS2MCP
MIT ★ 2 -
Give AI coding agents your Figma design in a compact format: ~94% fewer tokens than raw Figma JSON. Figma plugin + Spring AI MCP server.
MIT -
Never lose context to compaction. Archive pre-compaction transcripts, restore working state, and let the model dig back into what it forgot — MCP tools for Claude Code.
MIT ★ 1 -
@asafelobotomy/ceeline-mcp-servergithub.com/asafelobotomy/ceelineupdated this year
MCP JSON-RPC server for Ceeline compact transport tools
npm MIT -
Regression-Aware Context Compaction & Witness Replay for Coding Agents (CEGAR-Harness)
MIT -
claude-praetorian-mcpgithub.com/vvkmnn/claude-praetorian-mcpupdated this year
An MCP server for context compaction and recycling in Claude Code
npm MIT -
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 -
Progressive Skill Disclosure & Context Compaction Engine for Autonomous AI Coding Agents
npm Apache-2.0 -
Context compaction and safety gating for AI agents via TypeSafe Jev: keeps messages verbatim, no summarization. OpenAI, Anthropic, LangChain, CLI, MCP.
MIT ★ 2 -
Free MCP stdio bridge for agent continuity: resume after compaction, hand off between runtimes, recover from failure. No API key.
npm MIT -
@mseep/zipmem-mcpgithub.com/ahmetakyurt/zipmem-mcpupdated this year
Local-first MCP server + CLI that gives terminal AI agents an endless semantic memory via Anchored Compacting, keeping the context window small.
npm MIT -
Free MCP bridge for agents that lose context: resume after compaction, hand off between runtimes, recover from failure. No API key, no account, no payment.
npm MIT -
Context optimization middleware for LLM agents — a dynamic tool registry, result masking, token budgeting and smart compaction, so a long session stays inside its window instead of failing for length.
npm MIT -
vMemgithub.com/soolaugust/vMemupdated this year
Zero context compaction for Claude Code & LLM agents. Persistent memory powered by OS primitives (demand paging, kswapd eviction, mlock pinning). Single SQLite file, MCP-native, multi-agent shared.
★ 5 -
Local decision ledger for AI coding agents — keeps why a choice was made and which paths are dead ends, and re-injects it after context compaction. Plain Markdown, zero dependencies, works across Claude Code / Codex / Cursor / Copilot / Gemini.
MIT -
claude-code-ctxgithub.com/tugayoktayokay/claude-code-ctxupdated this year
Claude Code context manager + memory engine. Primary install is as a Claude Code plugin (see README). Also usable standalone as a zero-dep CLI: token monitoring against a model-aware quality ceiling, tailored /compact prompt + auto-clipboard, JSONL gzip b
npm MIT -
memory-mcpgithub.com/o-kasian/memory-mcpupdated this year
Simple MCP for logging and compacting the memory
MIT -
Give agents fast, accurate, and compact context on your Python codebases
MIT ★ 1 -
MCP server: locale-aware number formatting (currency, percent, compact).
MIT -
Compact MCP/CLI for on-premises Azure DevOps
MIT -
writing-context-rtfmactive
Context engine prioritizing RTFM search results into compact context packs
pypi MIT -
Read-first Stripe MCP server in Rust with compact responses and guarded metadata writes
MIT -
EfficientSearchgithub.com/om-surushe/EfficientSearchupdated this year
MCP server for compact Google Programmable Search results.
MIT -
Single-file MCP server for Trello: compact replies, reversible writes only, stdio and HTTP transports
MIT -
studio-mcpgithub.com/studio-mcp/studio-mcpupdated this year
🌞 Turn any CLI command into a one-room MCP server. npx -y studio-mcp; Clean, compact. Move-in ready.
npm MIT -
Google News headlines for AI agents as compact markdown, over MCP. No API key.
-
Token-lean MCP server for Figma: read designs as compact text, render only on demand
npm MIT -
Context-Canarygithub.com/amarisaster/Context-Canaryupdated this year
A token counter overlay for Claude. Warns you before context compaction hits so you can save what matters.
Apache-2.0 ★ 5 -
errsgithub.com/hegner123/errsupdated this year
Compact error/lint parser MCP tool. Normalizes verbose compiler/linter output into structured JSON.
MIT -
MCP server that gives AI agents a compact symbol map of a codebase (Rust)
MIT ★ 1 -
vcm-mcpno update data
Token-aware compaction of .vcm/history.json into SKILL.md, plus read/write of SKILL.md. Runs locally; the summarizer call goes to any OpenAI-compatible endpoint.
smithery -
memocallgithub.com/baodq06/memocallupdated this year
An MCP server that lets Claude Code recall past conversations from any project, as clean compact Markdown.
npm MIT ★ 3 -
RouterOS 7 REST MCP server for AI agents — compact filtered reads, confirm-gated mutations, scheduler-armed auto-rollback
pypi MIT -
Measures how a webpage actually renders and prints it as compact text for AI coding agents.
npm MIT -
distilno update data
Context optimization middleware for LLM agents — a dynamic tool registry, result masking, token budgeting and smart compaction, so a long session stays inside its window instead of failing for length.
smithery -
Provider-neutral PostgreSQL + pgvector memory with Codex compaction continuity
MIT -
Lean, identifier-native MCP server for Vikunja: compact task payloads and project keys like INFRA-41 instead of opaque global IDs.
npm MIT -
claude-praetorian-mcpgithub.com/Vvkmnn/claude-praetorian-mcpupdated this year
⚜️ An MCP server for context compaction and recycling in Claude Code
npm MIT ★ 10 -
Turns build, test and runtime failures into compact, agent-ready repair briefs. Local-first VS Code extension.
MIT -
Your Figma selection as compact text an AI agent reads and edits: Figma plugin, local bridge and MCP server
npm MIT -
NodeLane Actno update data
Use Reddit and Zhihu through your signed-in local browser with three compact MCP tools. Requires Node.js 22+ and the companion browser extension on the same computer. No NodeLane cloud relay.
smithery -
Context continuity for Claude Code & Grok — structured handoffs keep agent memory intact across compaction.
NOASSERTION -
MCP server for Linux/Arch administration over SSH: 113 typed tools, compact JSON, containers, sudo, two-step confirmation
npm MIT ★ 2 -
project-graph-mcpgithub.com/rnd-pro/project-graph-mcpupdated this year
MCP server for AI agents with compact project graphs, code analysis, and browser-test context.
npm MIT ★ 3 -
mysql-context-mcpgithub.com/velinussage/mysql-context-mcpupdated this year
Local MySQL binlog indexer with a stdio MCP server. Agents pull compact field diffs (from/to), not live table rows. No UI.
MIT -
Causal memory layer for AI agents — MCP server that records decision→outcome relationships. Survives compaction.
pypi Apache-2.0 ★ 107 -
Agent emotional continuity with PAD state, trust, appraisal, and compact emotion logs.
MIT ★ 29 -
Node.js FHIR MCP server with SMART Backend Services, metadata-aware search/CRUD tools, compact responses, and FHIRPath filtering.
npm Kopimi -
MCP server and CLI that inspect and query large JSON and JSONL files with jq-like expressions - compact results, no context flooding.
MIT ★ 1 -
json_to_toon_compressor converts oversized JSON arrays into a compact toon-style textual representation that removes wasted JSON syntax while preserving structural meaning needed by agents.
-
Lode — keep AI coding on one reviewed product direction across Claude Code, Codex & OpenCode, with compaction-safe recovery and change lineage.
MIT ★ 2 -
PocketBase CLI for AI agents and people: compact output, full API coverage, and a read-only mode that can be enforced
MIT -
MCP server for TMDB with compact, agent-friendly tools: movies, TV, people, discover by genre and streaming service, where to watch. Stdio and authenticated HTTP.
MIT -
MCP server that exposes every Raijin tool to AI agents over stdio: one typed tool per binary, same compact JSON. Part of Raijin.
MIT -
repforgithub.com/hegner123/repforupdated this year
MCP server tool for safe, exact-match string replacement across files. Designed for AI-driven refactoring workflows with compact JSON output.
go MIT -
siggithub.com/hegner123/sigupdated this year
Extract public API surface from source files as compact JSON: signatures, types, interfaces, no bodies. Supports Go, TypeScript, C#. MCP server + CLI.
go MIT -
Local CLI and MCP server for compacting test and build output, with retrieval from a redacted local cache.
MIT ★ 43 -
Fast Rust Android automation for AI agents: CLI + MCP, compact screen reads, action batches and paired Wi-Fi accessibility control (preview).
NOASSERTION ★ 1 -
Local Trello MCP for Claude, Codex and ChatGPT: compact responses, workspace allowlists and read-only opt-in.
go GPL-3.0 ★ 5