Self-host engrim
v1.4.11Cross-agent episodic memory for AI coding assistants
- Stars
- 301
- Requires
- Python
What is engrim?
Engrim is a local-first memory layer for AI coding assistants that stores project decisions and context in a SQLite database on your machine. It works with several popular tools by injecting a compact memory pack at the start of each session. This lets you clear chat history without losing architectural knowledge. The system uses hybrid search combining full-text and vector embeddings for fast recall, tracks which agent created each memory, and provides an MCP server so agents can read and write memories directly. A doctor command checks database health and fixes broken hook paths automatically. All data remains local with strict file permissions and no telemetry. Installation is a single pip command requiring Python 3.10 or newer.
Should you self-host engrim?
engrim scores 48/100 on selfhost.directory (how scoring works) and ranks #92 of 166 in Developer Tools. It has 301 GitHub stars.
It shipped 9 stable releases in the past year. The latest, v1.4.2, came out on 23 September 2026.
Setup is easy: most people get it running in one sitting. It is MIT-licensed and written mostly in Python.
Its closest alternative on selfhost.directory is openhands (87/100, 92 086 stars). It scores higher overall, so compare the two before you pick. See engrim vs openhands side by side.
Features
- 100% local and offline
- Agent provenance tracking
- Continue-As-Clear workflow
- Cross-agent episodic memory
- Flight recorder logging
- Hybrid lexical + vector recall
- MCP server for tool exposure
- Memory merge and backup
- Resume pointer workflow
- Safe clear review
- Self-healing hook diagnostics
engrim README
More in Local AI
exllamaOpen-source Python/C++/CUDA runner for 4-bit GPTQ-quantized Llama models on consumer NVIDIA GPUs, with a self-hosted chat web UI you can build and run in Docker.
exoSelf-hosted, open-source AI cluster that links your Macs and Linux boxes to run large LLMs locally via MLX, with Thunderbolt 5 RDMA and an OpenAI/Ollama-compatible API.
FastChatSelf-hosted, open-source LLM serving platform: run open chat models behind a Gradio web UI, a CLI, or an OpenAI-compatible REST API.
Home AssistantSelf-hosted open-source home automation platform with 1500+ integrations, powerful automations, customizable dashboards, and a built-in voice assistant.
homelabSelf-hosted autonomous agent factory running OpenClaw, Kilo Pipeline, Crawl4AI, Qdrant, and Ollama locally on low-power x86 hardware.
InvokeAISelf-hosted, open-source AI image generation studio with a layered Unified Canvas, node-based workflows, ControlNet, and broad local model support from SD 1.5 to Flux.
jevosOpen-source, self-hosted decision model answering yes/no, choice and score questions with probabilities on a laptop CPU in 28-130 ms, an alternative to Jev's cloud.
koboldcppSelf-hosted, open-source (AGPL-3.0) single-file GGUF LLM server with a bundled KoboldAI Lite web UI and OpenAI-, KoboldAI- and Ollama-compatible API endpoints.
lemonadeSelf-hosted, open-source local AI server that runs LLM, vision, image, speech, and transcription models on your own GPU or NPU and serves them through OpenAI-style APIs.
llama.cppllama.cpp is an open-source, self-hosted LLM inference engine written in C/C++ that runs quantized models locally on CPU and GPU with minimal dependencies.
llama-gptSelf-hosted, open-source ChatGPT-style chatbot that runs Llama 2 and Code Llama locally, with a web chat UI and an OpenAI-compatible API.
llmakerllmaker is a self-hosted, open-source CLI that provisions and manages complete LLM stacks on your own hardware with a single command.
0 comments