nanoclaw-repl

Manage NanoClaw v2 sessions with markdown-backed commands and branching.

2|Updated Mar 20, 2026
One-click install
npx skills add https://github.com/multiplex-ai/muggle-ai-teams --skill nanoclaw-repl-multiplex-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: nanoclaw-repl
Source: https://github.com/multiplex-ai/muggle-ai-teams/tree/main/skills/nanoclaw-repl
Command: npx skills add https://github.com/multiplex-ai/muggle-ai-teams --skill nanoclaw-repl-multiplex-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

NanoClaw REPL provides a compact, zero-dependency environment to operate and extend NanoClaw v2 sessions inside Claude Code, enabling persistent session state and interactive experimentation without external runtimes.

Core Features & Use Cases

  • Persistent markdown-backed sessions
  • model switching with /model
  • dynamic skill loading with /load
  • session branching with /branch
  • cross-session search with /search
  • history compaction with /compact
  • export to md/json/txt with /export
  • session metrics with /metrics

Use cases include rapidly prototyping new commands, debugging session flows, and maintaining long-running interactive tasks within a Claude Code session.

Quick Start

Launch nanoclaw-repl in Claude Code and connect to a NanoClaw v2 session to begin interactive development.

Frequently Asked Questions about nanoclaw-repl

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I manage persistent session state in Claude Code without external runtime dependencies?

NanoClaw REPL enables persistent, markdown-backed session state management inside Claude Code without requiring external runtimes. It provides a zero-dependency environment for maintaining long-running interactive tasks directly within your repository.

How do I switch models or branch sessions during an interactive Claude Code task?

You can switch models with the /model command and create per-task session branches using /branch. These deterministic command handlers allow dynamic model switching and session branching within NanoClaw v2 workflows.

Can I load new skills dynamically into a running NanoClaw v2 session?

Yes, you can dynamically load skills into a running session using the /load command. This enables rapid prototyping and in-repo extension of commands without restarting the interactive Claude Code environment.

What is the best way to search across markdown-backed interactive sessions and export history?

You can perform cross-session search with /search and export history to md, json, or txt formats using /export. The /compact command also compacts history to manage session metrics efficiently.

Do I need external libraries or frameworks to operate a deterministic REPL for Claude Code?

No external libraries are required. NanoClaw REPL operates as a zero-dependency skill, utilizing deterministic command handlers and markdown-backed sessions to manage interactive workflows directly inside Claude Code.