
Nanoclaw Repl
- 1.4k installs
- 238k repo stars
- Updated August 5, 2026
- affaan-m/ecc
This is a copy of nanoclaw-repl by affaan-m - installs and ranking accrue to the original listing.
nanoclaw-repl is a Claude Code skill that runs persistent, branching REPL sessions for developers who need task-focused state across extended Claude-powered agent workflows.
About
nanoclaw-repl is an ECC-origin skill for running persistent, branching REPL sessions that keep Claude agent work on-task across multiple turns and branches. It targets developers extending agent workflows who want REPL-like continuity instead of stateless prompt chains. Sessions remain focused on the active task while supporting branch exploration for alternatives or parallel hypotheses. The published readme excerpt is a Japanese translation stub pointing to the English SKILL.md source, so behavior is defined by the skill metadata and ECC agent patterns rather than the stub file alone.
- Persistent markdown-backed sessions that survive across restarts
- 8 core commands including /model, /load, /branch, /search, /compact, /export and /metrics
- Session branching before high-risk changes with one command
- History compaction after major milestones to maintain context quality
- Zero external runtime dependencies with deterministic local command handlers
Nanoclaw Repl by the numbers
- 1,357 all-time installs (skills.sh)
- +83 installs in the week ending Aug 5, 2026 (Skillselion tracking)
- Data as of Aug 5, 2026 (Skillselion catalog sync)
npx skills add https://github.com/affaan-m/ecc --skill nanoclaw-replAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 1.4k |
|---|---|
| repo stars | ★ 238k |
| Last updated | August 5, 2026 |
| Repository | affaan-m/ecc ↗ |
How do you keep state in agent REPL sessions?
Run persistent, branching REPL sessions that stay task-focused while extending Claude-powered agent workflows.
Who is it for?
Developers building multi-step Claude agent workflows who need REPL persistence and branch exploration without losing task focus.
Skip if: Developers who only need one-shot prompts or production observability should use simpler invoke patterns or monitoring skills instead.
When should I use this skill?
The user wants a persistent REPL, branching agent session, or long-running task-focused Claude workflow extension.
What you get
Persistent REPL session branches, retained task context, and extended Claude agent workflow runs.
- Persistent REPL session state
- Branching agent workflow runs
Files
nanoclaw-repl - 日本語翻訳進行中
このファイルの翻訳は実装中です。英語版は元のスキルファイルを参照してください。
詳細は:D:/tmp/everything-claude-code/skills/nanoclaw-repl/SKILL.md
Related skills
How it compares
Use nanoclaw-repl for interactive agent REPL persistence; use performance-monitor skills when the goal is metrics and SLA tracking instead of session continuity.
FAQ
What is nanoclaw-repl for?
nanoclaw-repl is an ECC skill that runs persistent, branching REPL sessions for Claude agent workflows. It keeps work task-focused while letting developers explore branches without resetting conversational and execution context.
Does nanoclaw-repl replace a normal terminal REPL?
nanoclaw-repl layers agent-oriented persistence and branching on Claude workflows inside ECC. It complements—not replaces—a system shell, targeting task-focused agent sessions rather than general command-line administration.