multi-agent-ralph-loop
Multi-agent coding orchestration with memory, quality gates, and security
All Skills in This Repository (33)
Pure Emerald Level Indicatorsskill
Create, validate, and orchestrate YAML-based skills in Ralph's ecosystem.
kaizen
Plan, execute, and review structured continuous improvement cycles for teams.
context7-usage
Resolve library IDs and query Context7 documentation with fallback to MiniMax or WebSearch.
gates
Run linting, formatting, type checking, and tests across multiple programming languages.
codex-cli
Orchestrate OpenAI Codex CLI tasks for multi-agent software development.
sec-context-depth
Identify security vulnerabilities in AI-generated code using 27 sec-context anti-patterns.
parallel
Coordinate parallel Ralph loops with isolated state and aggregate results.
compact
Save conversation ledgers and handoffs manually before starting fresh sessions.
stop-slop
Remove AI tells and writing slop from prose while preserving meaning and voice.
testing-anti-patterns
Identify and document testing anti-patterns in Claude-style skills.
security
Identify security vulnerabilities and generate JSON findings with CWE and OWASP references.
clarify
Plan and execute structured requirement clarification using the AskUserQuestion workflow.
Frequently Asked Questions
FAQPage SchemaHow to install Multi-Agent Ralph Loop?â–¼
Run `npx skills add alfredolopez80/multi-agent-ralph-loop --all -g -y` in your terminal to install all skills globally.
What does Multi-Agent Ralph Loop do?â–¼
It turns a single coding session into a coordinated team of specialized agents that plan, implement, review, test, and security-audit your code. Every task must pass blocking quality gates before it is marked done.
How do I run a task with multiple agents?â–¼
Invoke the /orchestrator skill with your task description, and it will clarify requirements, plan the work, spawn coder, reviewer, and tester teammates, and validate the result automatically.
Can it run a whole backlog of tasks autonomously?â–¼
Yes. The /task-batch skill reads a PRD or task list and executes every task with fresh context, auto-commits, and completion criteria until all tasks are verified done.
Does it work without deep coding knowledge?â–¼
You still need a development project to work on, but you describe goals in plain English and the agents handle implementation, testing, and security checks for you.
Related Repositories in Software Engineering
View All in Software Engineering→openclaw
Run a personal AI assistant across your devices and chat apps
superpowers
Gives coding agents a disciplined workflow from idea to merged code
react
AI agent skills for building, testing, and porting React core