Nanoclaw Repl
A session-aware REPL built on top of `claude -p` that stores everything in markdown files instead of a database. The zero-dependency architecture is deliberate and actually matters here since it runs as a Node script without external runtime deps. You get model switching, session branching, and cross-session search, all persisted to markdown that you can read and edit directly. The workflow is meant for iterative development where you might want to branch before trying something risky or compact your history after hitting a milestone. It's opinionated about staying local and deterministic, which makes sense for a REPL you're extending yourself. If you're already using Claude API and want conversation persistence without standing up infrastructure, this does that.
- installs
- 4.7K
- GitHub stars
- 229.1K
- last updated
- Jul 14, 2026
Install
$ npx -y skills add affaan-m/everything-claude-code --skill nanoclaw-repl --agent claude-codeInstalls into .claude/skills of the current project.
Files
NanoClaw REPL
Use this skill when running or extending scripts/claw.js.
Capabilities
- 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
Operating Guidance
- Keep sessions task-focused.
- Branch before high-risk changes.
- Compact after major milestones.
- Export before sharing or archival.
Extension Rules
- keep zero external runtime dependencies
- preserve markdown-as-database compatibility
- keep command handlers deterministic and local
Sponsors
