Mnemo AI
A local agentic AI assistant with MCP (Model Context Protocol) integration, RAG capabilities, and intelligent conversation management. Built on LangGraph with LangChain for multi-provider LLM support (Ollama, MLX, Amazon Bedrock, Bedrock Mantle, OpenAI, Anthropic, Amazon SageMaker AI, LiteLLM).
📖 Documentation
Full documentation is available at https://brunopistone.github.io/mnemoai/
- Getting Started
- Usage & commands
- Configuration
- Orchestration & sub-agents
- Memory & learning
- Everyday tools
- Architecture
- Development
🚀 Quick Start
pip install mnemoai-assistant # or: uv tool install mnemoai-assistant
mnemoai # verbose (shows thinking); --no-verbose to hide
On first run, if no config is found, an interactive configurator launches and walks you through picking a provider, model, and feature toggles — then writes ~/.mnemoai/config/config.yaml.
→ See the Getting Started guide for full setup.
✨ Key Features
- 🤖 Multi-Model Support: Ollama (local), MLX (local, Apple Silicon), Amazon Bedrock, Bedrock Mantle, OpenAI, Anthropic (Claude), Amazon SageMaker AI (chat only — no tool calling), LiteLLM (100+ providers)
- 🔧 MCP Tool System: Extensible tool architecture via Model Context Protocol
- 📚 RAG: Automatic document indexing and semantic (hybrid) search
- 🧠 User Profile Learning: Personalized responses learned from interactions
- 🧩 Episodic Memory: Learns from successful task completions and retrieves similar solutions
- 📖 ACE Playbook: Learns strategies from successes AND failures (Agentic Context Engineering)
- 🔍 Web Search & 🌐 Crawler: Brave Search API + web page extraction with RAG ingestion
- 🖼️ Vision Support: Image analysis with vision models
- 📁 File Operations & ✏️ Precise Editing: Read/write/edit text, CSV, JSON, PDF, DOCX
- 🔎 Fast Search: Glob + ripgrep content search (10-100x faster)
- 📋 Todo Tracking, 📝 Plan Mode & 🔄 Background Tasks: Multi-step task management
- ⚡ Bash Execution & 🛡️ Git Safety: Shell commands with smart error handling and guardrails
- 🪝 Tool Hooks: Your own commands run before/after any tool call — format writes, deny paths, auto-approve safe commands (docs)
- ⌨️ Your Own Slash Commands: A markdown file per prompt you retype —
commands/review.mdbecomes/review <path>(docs) - 📎
@-File Mentions: Type@to complete a path anywhere in your prompt; the file is read and sent with the question (docs) - ⟲ Take Back a Prompt:
/rewinddrops your last prompt and everything the turn produced — the conversation only, files on disk untouched (docs) - 🗂️ Workspace Reports:
/fileslists what the session read, changed or attached;/diffshows uncommitted changes with this session's edits marked;/copyputs the last answer on the clipboard without the terminal's wrapping (docs)
📄 License
Licensed under the MIT License — see the LICENSE file for details.
🤝 Contributing
This is a personal development project. Feel free to fork and adapt it to your needs; attribution to the original repository is appreciated but not required.
1 maintainer
Release files for mnemoai-assistant 1.20.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| mnemoai_assistant-1.20.0.tar.gz | 2.3 MB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| mnemoai_assistant-1.20.0-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 3.0 MB
Release files / mnemoai_assistant-1.20.0.tar.gz
| Download URL | mnemoai_assistant-1.20.0.tar.gz |
|---|---|
| Size | 2.3 MB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
af93d7dda3532e071ce1bd6d2b432c165268a5f29ae8e284490171e6f418c082
|
|
BLAKE2b-256 checksum How to use checksums |
77f665881fc5620b4a23f71ab881beae9cb1620e6bea49dae5ccfe6c415de6e1
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Release files / mnemoai_assistant-1.20.0-py3-none-any.whl
| Download URL | mnemoai_assistant-1.20.0-py3-none-any.whl |
|---|---|
| Size | 626.6 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
fda400f1bc70f55364783ef6bc9ed530da7f7895abd292449708288eb44ebda7
|
|
BLAKE2b-256 checksum How to use checksums |
5d388df402335bc3a58413f1e8e62481c9adb74fb3fa7ea7f5c35b84b633296e
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
