MCP servers for local llm

The first 20 by name of 60 servers in the official MCP Registry (registry.modelcontextprotocol.io) that match “local llm”.

Servers in the official MCP Registry (registry.modelcontextprotocol.io) whose name or description contains every word of “local llm”, latest version of each, sorted by name; at most the first 20 are shown. The registry publishes no popularity count. Descriptions are the authors' own words. Data fetched 2026-11-20.

ServerByDescription
agentmailkit
pypi
io.github.ariaxhan/agentmailkit“Run scheduled LLM-written email digests from local files and feeds. Local-first MCP server.”
Algernon MCP
pypi
io.github.sammyboi81/algernon“Fan a goal out to cheap parallel LLM workers on your own key or a free local Ollama. Keep your mind.”
argot
npm
io.github.get-tmonier/argot“Read-only context and complete checks against patterns learned from git history. Local; no LLM.”
Awareness Local
npm
io.github.everest-an/awareness-local“Local-first persistent memory for AI coding agents - 96.0% R@5 on LongMemEval, zero LLM calls.”
Chat Export
oci
io.github.svergio/chat-export-server“Local MCP for eight logged-in LLM UIs via the Chat Export Chrome extension on 127.0.0.1.”
ChimeraForge
pypi
io.github.Sahil170595/chimeraforge“Local-first LLM deployment planner: GPU/VRAM sizing, cost and latency, with provenance”
cogvault
pypi
io.github.NBibikov/cogvault“Local hybrid-search memory for AI agents over plain Markdown. Multi-tenant, offline, no LLM.”
com.runmemento/memento
npm
com.runmemento/memento“Local-first, LLM-agnostic memory layer for AI assistants.”
context-doctor
npm
io.github.KushalP1/context-doctor“Cut LLM agent token use in Claude Code, Cursor, Codex, Gemini CLI. Local, no API key. gAI Ventures.”
dead-letter
mcpb, oci, pypi
io.github.BigCactusLabs/dead-letter“Convert .eml and .mbox email to clean Markdown for RAG, LLM pipelines, and local knowledge bases”
Dense Knowledge
pypi
io.github.Lucky44k/dense-knowledge-mcp“Local-first MCP memory server for persistent LLM knowledge with BM25 retrieval.”
Docpick
pypi
io.github.ArkNill/docpick“Schema-driven document extraction with local OCR + LLM. Document in, Structured JSON out.”
ellmos Homebase
npm
io.github.ellmos-ai/ellmos-homebase-mcp“Local-first LLM orchestration: memory, knowledge, state, routing, swarm, API discovery, plugins.”
Engram
pypi
io.github.TushaeBXN/engram“Local, private long-term memory for AI: verbatim, searchable, correctable. No LLM or API key.”
evalpair
pypi
io.github.Srikarmk/evalpair“Run LLM eval suites on local or hosted models; compare runs with intervals and a noise floor.”
Fuuga minion
nuget
io.github.Thorium/fuuga“Offline local-LLM helper: delegate simple errands to a local model, with sandboxed file tools.”
HarnessScope
npm, pypi
io.github.moongioh/hscope“Turn-level governance observability for Claude Code transcripts. Local-first, no LLM inside.”
io.github.aapd-studio/omnideck
npm
io.github.aapd-studio/omnideck“Local privacy-first tools: redact secrets, decode JWTs, compare LLM costs. No network calls.”
io.github.Abidit/phi-guard-mcp
npm
io.github.Abidit/phi-guard-mcp“Local-first PHI detection for source code, LLM prompts, logs, and analytics”
io.github.base76-research-lab/token-compressor
pypi
io.github.base76-research-lab/token-compressor“Compress prompts 40-60% using local LLM + embedding validation. Preserves all conditionals.”

Data: MCP Registry API, fetched 2026-11-20. Page built 2026-11-21 · MCP servers for every need: 329 lists from the official MCP Registry