moltis and clawhive
These are competitors: both provide sandboxed, Rust-native agent runtimes with built-in tool execution, but Moltis offers more mature features (voice, memory, MCP protocol support) while Clawhive is an earlier-stage alternative in the same problem space.
About moltis
moltis-org/moltis
A Rust-native claw you can trust. One binary — sandboxed, secure, auditable. Voice, memory, MCP tools, and multi-channel access built-in.
Moltis helps you run a secure, private AI assistant on your own hardware, like a Mac Mini or Raspberry Pi. It takes your commands via voice, chat (Telegram, Discord), or web interface, processes them with various AI models, and produces automated actions, scheduled tasks, or responses. This is ideal for individuals or small teams who want a powerful, customizable, and secure AI agent without relying on third-party cloud services for data processing or storage.
About clawhive
longzhi/clawhive
Lightweight, Rust-native AI Agent platform. Security sandbox from day one.
Supports multi-agent orchestration with per-agent personas and model routing, three-layer memory system (session JSONL → daily files → long-term synthesis), and hybrid search combining vector similarity with BM25 full-text indexing. Integrates across eight chat platforms (Telegram, Discord, Slack, WhatsApp, iMessage, Feishu, DingTalk, WeCom) through pluggable channel adapters, with a two-layer security model enforcing hard baseline restrictions plus permission-based sandboxing for external skills.
Scores updated daily from GitHub, PyPI, and npm data. How scores work