Key takeaways
- Single agent with unified memory across 5 channels — Telegram, Slack, Discord, WhatsApp, Signal
- Persistent memory across days/weeks/months, not just sessions
- Built on Letta platform (formerly MemGPT) — advanced memory and self-improvement
- Compatible with ClawHub and skills.sh for extensibility
FAQ
How is LettaBot different from OpenClaw?
LettaBot focuses on memory — one agent that remembers everything across all channels and sessions. OpenClaw has more features but simpler memory. LettaBot is built on Letta's stateful agent platform.
What makes Letta's memory special?
Letta (formerly MemGPT) pioneered stateful agents with persistent memory that can learn and self-improve. Memory persists across days, weeks, and months — not just a single session.
Who built LettaBot?
Letta AI, the company behind MemGPT and Letta Code. YC-backed, focused on AI with advanced memory.
Overview
LettaBot is a personal AI assistant built on the Letta platform (formerly MemGPT), focused on persistent memory across multiple channels.[1] Unlike most AI assistants that forget between sessions, LettaBot maintains a single unified memory across Telegram, Slack, Discord, WhatsApp, and Signal — remembering conversations from days, weeks, or months ago.
Built by the Letta AI team, it represents the "memory-first" approach to personal AI assistants.
Key Stats
| Metric | Value |
|---|---|
| Platform | Letta (formerly MemGPT) |
| Language | TypeScript |
| Channels | Telegram, Slack, Discord, WhatsApp, Signal |
| License | Apache-2.0 |
| Skills | ClawHub + skills.sh compatible |
Architecture
Telegram ──┐
Slack ─────┤
Discord ───┼──→ ONE AGENT ──→ ONE CONVERSATION
WhatsApp ──┤ (memory) (chat history)
Signal ────┘
Single agent, unified memory. Start on Telegram, continue on Slack, pick up on WhatsApp — the agent remembers everything.
Features
- Multi-Channel — Chat across 5 platforms with unified context
- Persistent Memory — Remembers across days/weeks/months, not just sessions
- Voice Messages — OpenAI Whisper transcription across all channels
- Heartbeat — Periodic check-ins where agent reviews tasks
- Scheduling — One-off reminders and recurring tasks
- Local Tool Execution — Read files, search code, run commands
- Streaming Responses — Real-time updates as agent thinks
- Skills — ClawHub and skills.sh compatible
Installation
git clone https://github.com/letta-ai/lettabot.git
cd lettabot
npm install && npm run build && npm link
lettabot onboard
lettabot server
Requires Node.js 20+ and a Letta API key from app.letta.com.
The Letta Platform
LettaBot is built on Letta, the platform for stateful agents.[2] Letta (formerly MemGPT) pioneered the concept of AI with advanced memory that can learn and self-improve over time.
Key Letta features:
- Context Repositories — Git-based memory for coding agents
- Conversations API — Shared memory across concurrent experiences
- Letta Code — #1 model-agnostic agent on Terminal-Bench
- Self-improvement — Agents that learn from their mistakes
Security
| Feature | Implementation |
|---|---|
| Network | Outbound connections only, no exposed ports |
| Tool Execution | Read-only by default (read, glob, grep, search) |
| Access Control | Pairing-based — users need approval codes |
| Architecture | No public URL or gateway required |
Strengths
- Memory-First — Persistent memory is the core feature, not an afterthought
- Unified Context — One agent across all channels
- Platform Backing — Built on mature Letta infrastructure
- Voice Support — Whisper transcription across all channels
- Skills Ecosystem — ClawHub + skills.sh compatibility
- Self-Hosted — Run on your own infrastructure
Weaknesses / Considerations
- Letta Dependency — Requires Letta API key or self-hosted server
- Complexity — More setup than simpler alternatives
- Node.js Required — Not as lightweight as Go/Rust options
- Newer Project — Less community than OpenClaw
Who Built It
Letta AI, the company behind MemGPT. YC-backed, focused on AI with advanced memory that can learn and self-improve over time. The team includes researchers who pioneered the MemGPT concept.
Ideal User
- Memory-focused users who need persistent context across sessions
- Multi-platform users chatting across Telegram, Slack, Discord, WhatsApp, Signal
- Letta ecosystem users already using Letta Code or Letta API
- Developers wanting a hackable, skills-compatible assistant
Bottom Line
LettaBot is what you run when memory matters more than features. While OpenClaw has 50+ integrations and extensive capabilities, LettaBot focuses on one thing: remembering everything across all channels and sessions.
The unified memory architecture — one agent, one conversation across 5 platforms — is genuinely different from most multi-channel bots that maintain separate contexts. For users who want their assistant to actually know them over time, this is the most memory-focused option available.
The trade-off is ecosystem size. OpenClaw has more features, more integrations, more community. But if persistent memory is your priority, LettaBot is purpose-built for it.