Cc Mini — AI skill for Claude Code
Ultra-light Harness scaffolding for AI agents, a mini version of claude code.
How to install Cc Mini
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open e10nMa2k/cc-mini and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Cc Mini does
Ultra-light Harness scaffolding for AI agents, a mini version of claude code.
Alternatives in AI
- PRD: Agent Builder — Create Your Own Sub-Agent — Your ecosystem, your rules 1.9k ★
- Claude Fable 5 System Prompt Clean — the optimized, token-efficient version of the leaked Claude Fable 5 / Mythos 5 system prompt 475 ★
- SAP Skills - Project Context — Repository: https://github.com/secondsky/sap-skills Purpose: Production-ready skills for SAP development and A 207 ★
README
cc-mini
**Ultra-light Harness scaffolding for AI agents**
**Agentic** · **Built to Extend** · **From Claude Code**
The entire core is `~1000 lines of Python`
**NEW: Buddy — AI Companion with Custom Sprites**
Your coding companion lives in the terminal. Type `/buddy` to hatch it. Supports custom ASCII species — bring your own Pikachu!

[Full Buddy docs →](docs/buddy.md)
Features
Core
- Interactive REPL with streaming output, command history, slash command autocomplete
- Agentic tool loop — Claude calls tools autonomously until the task is complete
- 9 built-in tools:
Read,Edit,Write,Glob,Grep,Bash,AskUser,EnterPlanMode,ExitPlanMode - Plan mode — parallel subagents explore codebase before you implement, with permission isolation
- Permission system — mode-aware (default/plan), reads auto-approved, writes/bash ask for confirmation
- Session persistence — auto-save conversations,
/resumeto continue later - Context compression — auto-compact when approaching token limits
- Anthropic + OpenAI compatible — works with any compatible API endpoint
Advanced (from unreleased Claude Code features)
| Feature | Description | Docs |
|---|---|---|
| Coordinator Mode | Background workers for parallel research and implementation | docs → |
| Buddy | Tamagotchi AI pet with personality, stats, mood, and speech bubbles | docs → |
| KAIROS Memory | Cross-session memory with auto-consolidation | docs → |
| Skills | One-command workflows: /review, /commit, /test, /simplify |
docs → |
| Sandbox | Bubblewrap isolation for bash commands | docs → |
Quick Start
Requirements
- Python 3.10+ (3.11+
Related Skills
Ground Truth Evals
An LLM eval harness graded by computed ground truth, not an LLM judge. Worked example: poker, served to models
Openclaw Ultra Scraping
🕷️ Adaptive web scraping skill for OpenClaw agents — bypasses anti-bot, survives site redesigns. Powered by M
Fast File Search
Ultra-fast file search toolkit for AI agents, Neovim, and code editors. MCP server for Claude Code/Cursor/Code
Zeptoclaw Skills
Community skills for ZeptoClaw — the ultra-lightweight AI agent runtime
Micro Actinf
⚡ Ultra-fast O(1) Active Inference & Bayesian Cognitive Governor for AI Agents (Antigravity, Claude, Cursor) &
Ultra Scraper
🕷️ Agent-native web scraping library — stealth anti-bot bypass, auto-escalation, and native tool calling for
Related Agents
Package Maintainer
Use this agent for anything in ~/Packages — scaffolding new Laravel packages, upgrading dependencies or Larave
Harness Dev Ops
DevOps engineer — infrastructure, CI/CD, build tooling, deployment, environment config, scaffolding, dependenc
Caveman Compress
Compression engine wrapper — runs the caveman-compress CLI on a body artifact in scratch, returns the compress