Mnemolis
Active·★ 20·MIT·Updated 2026-10-16
Mnemolis is a unified local knowledge search API designed for self-hosted homelabs, routing queries to various internal backends like offline knowledge bases, weather forecasts, RSS feeds, and smart home states via a single endpoint. It offers both REST and MCP APIs, enabling natural language queries across diverse data sources entirely within your private infrastructure without sending data to third parties.
#Local Knowledge Base#API Gateway#Docker#LLM Routing#Home Automation#Multi-Source Search#Homelab#Privacy-Preserving
01
Features
01Unified API for various local knowledge sources (Kiwix, RSS, Weather, Home Assistant, Uptime Kuma, SearXNG).
02LLM-assisted intelligent routing, source selection, and query expansion.
03Multi-source concurrent query fusion and result merging.
04Snapshot engine to detect and report changes across services (outages, weather shifts, new headlines).
05Local-first and privacy-preserving design, with all major backends self-hosted.
02
Compatibility
Docker
Deployment
Verified via docs
Python
Runtime
Verified via docs
REST API
Interface
Verified via docs
MCP Client
Interface
Verified via docs
Ollama
LLM Backend
Verified via docs
OpenAI Compatible API
LLM Backend
Verified via docs
03
Quick start
1
$ git clone https://github.com/immortalbob/Mnemolis
2
$ cd Mnemolis
3
$ docker network create mnemo-net
4
$ cp docker-compose.example.yml docker-compose.yml
5
$ docker compose up -d
04
Use cases
↳Querying your self-hosted Wikipedia (Kiwix) with natural language.
↳Getting a combined summary of homelab service status and local weather conditions.
↳Asking about recent news from your private RSS feeds or current Home Assistant entity states.
↳Integrating with LLM clients like Claude Desktop or Open WebUI for enhanced conversational AI in a homelab.
↳Detecting and querying significant changes in your homelab environment, like service outages or temperature shifts.
05
Alternatives
obsidian-mcp-server★ 19
MCP server for Obsidian vault management - enables Claude and other AI assistants to read, write, search, and organize your notes
docker-claudebox★ 23
Claude Code in Docker. Drop-in OpenAI-compatible API, MCP server, Telegram bot, and CLI — five interfaces, one image. Persistent sessions, file ops, always-on skill injection, and a full dev toolchain (Go, Python, Node, K8s, Terraform, databases) or a minimal image with just the basics.
MCP-Bridge★ 923
A middleware to provide an openAI compatible endpoint that can call MCP tools
HasteHealth★ 84
Headless EHR for storing clinical data as FHIR R4 and serving it to apps and AI agents.
flexible-graphrag★ 194
Flexible GraphRAG: Python, LlamaIndex, Docker Compose: 8 Graph dbs, 10 Vector dbs, OpenSearch, Elasticsearch, Alfresco. 13 data sources (9 auto-sync), KG auto-building, schemas, LLMs, Docling or LlamaParse doc processing, GraphRAG, RAG only, Hybrid search, AI chat. React, Vue, Angular frontends, FastAPI backend, REST API, MCP Server. Please 🌟 Star
Comments
Log in to leave a comment
No comments yet. Be the first!