By nbiish
An autonomous agent loop plugin for Claude Code that implements the Ralph Loop pattern. It takes a prompt, plans tasks, and iteratively executes them using fresh contexts until completion.
Restart YOLO loop iteration (reset to 0)
Guide active YOLO session with new instructions
Execute task with Mini-SWE-Agent
Stop YOLO Mode completely (disable the autonomous loop)
Start YOLO Mode with TTS feedback
Runs pre-commands
Contains inline bash commands via ! syntax
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Transform Claude Code into a self-driving developer
YOLO Mode implements the Ralph Loop pattern for autonomous agentic coding. It plans, executes, and verifies complex tasks with minimal human intervention while maintaining context hygiene through fresh context windows.
YOLO_PLAN.md to track progresstts-cli--dangerously-skip-permissions for maximum autonomyIn Claude Code, add the marketplace and install the plugin:
/plugin marketplace add nbiish/yolo-mode
/plugin install yolo-mode@yolo-marketplace
If you install/enable/disable plugins during a session, run:
/reload-plugins
Use in Claude Code:
/yolo "Refactor the authentication system"
/yolo "Audit security" --agent opencode
/yolo-tts "Build a React component library" # With voice feedback
/yolo-guide "Focus on small diffs, add tests, and keep changes minimal"
/yolo-stop
Plugin Surface Area (what should appear in Claude Code):
yolo, yolo-tts/yolo <goal> - Start YOLO loop/yolo-tts <goal> - Start YOLO loop with TTS/yolo-mini <task> - Run a single task via Mini-SWE-Agent/yolo-guide <feedback> - Queue guidance for the next iteration/yolo-stop - Stop the loop and clear state/cancel-yolo - Hidden; resets iteration counter to 0# Install from source
git clone https://github.com/nbiish/yolo-mode.git
cd yolo-mode
pip install -e .
# Run anywhere
yolo-mode "Your goal here" --tts
yolo-mode "Your goal here" --agent opencode
/yolo "Create a REST API with FastAPI"
/yolo-tts "Write unit tests for the utils module"
# Basic usage (defaults to Claude Code)
yolo-mode "Implement user authentication"
# With OpenCode
yolo-mode "Refactor database schema" --agent opencode
# With Google Gemini
yolo-mode "Generate documentation" --agent gemini
# With Mini-SWE-Agent (NEW in v0.2.0)
yolo-mode "Write a sudoku game" --agent mini
/yolo-mini "Create unit tests for utils.py" # Slash command
# With voice feedback
yolo-mode "Build a dashboard" --tts
# Complex multi-step goal
yolo-mode "Set up a CI/CD pipeline with GitHub Actions, Docker, and AWS deployment"
npm install -g @anthropic-ai/claude-codebrew install opencodepip install mini-swe-agent (NEW in v0.2.0)YOLO Mode assumes your preferred provider + default model are already configured for each CLI tool (Claude/Gemini/OpenCode/etc.). When you can select a model (via a CLI flag or provider config), prefer:
deepseek/deepseek-v4-pro (flagship, tuned for agentic coding). Source: https://zenmux.ai/provider/deepseeknpx claudepluginhub nbiish/yolo-mode --plugin yolo-modeDescribe your goal, approve the spec, then step away — Claude and Codex loop together until it's right.
OpenAgentsControl — multi-agent orchestration for Claude Code. Context-aware development with skills, subagents, parallel execution, and automated code review.
This skill should be used when the model's ROLE_TYPE is orchestrator and needs to delegate tasks to specialist sub-agents. Provides scientific delegation framework ensuring world-building context (WHERE, WHAT, WHY) while preserving agent autonomy in implementation decisions (HOW). Use when planning task delegation, structuring sub-agent prompts, or coordinating multi-agent workflows.
Autonomous session orchestrator for Claude Code - manages multi-phase development workflows
Autonomous multi-agent development framework with spec-driven sprints and convergent iteration
Use Claude Code like a Director, not a Programmer. Complete toolkit with 26 commands, 14 agents, 31 skills, and TDD-based Auto-Loop.