skills.rest
xianmingyao
xianmingyaoCOMMUNITY
Explore all skills by this author →
GitHub

ralph-loop

Coordinate Ralph Loop workflows across coding agents with real TTY sessions.

1|1|Updated Mar 18, 2026
Search Tags:#ai-agent#coding-agent#tty#ralph-loop#exec-tool#process-tool
One-click install
npx skills add https://github.com/xianmingyao/openclaw-CaySon --skill ralph-loop-xianmingyao
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ralph-loop
Source: https://github.com/xianmingyao/openclaw-CaySon/tree/main/skills/ralph-loop-agent
Command: npx skills add https://github.com/xianmingyao/openclaw-CaySon --skill ralph-loop-xianmingyao

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Ralph Loop orchestration for OpenClaw agents, enabling end-to-end planning and building loops with real TTY support and structured progress tracking.

Core Features & Use Cases

  • Orchestrates OpenClaw coding agents (OpenCode, Codex, Claude Code, Goose) using exec with pty to provide a live terminal
  • Monitors long-running sessions with the process tool and checks completion against IMPLEMENTATION_PLAN.md
  • Supports PLANNING and BUILDING modes, backpressure iteration control, sandboxing, and safe work-dir isolation

Quick Start

Ask your OpenClaw agent to run a Ralph Loop end-to-end using exec and process tools with TTY support.

Frequently Asked Questions about ralph-loop

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I orchestrate interactive coding agents with real TTY sessions?▼

Orchestrate interactive coding agents by driving them through exec with pty to provide a live terminal, enabling real TTY sessions for tools like OpenCode, Codex, Claude Code, and Goose.

How do I monitor progress and completion of long-running coding agent workflows?▼

Monitor long-running coding agent workflows using the process tool to check session status and verify completion against the structured milestones persisted in IMPLEMENTATION_PLAN.md.

Does this orchestration approach support both planning and building modes?▼

Yes, the orchestration approach supports both PLANNING and BUILDING modes, allowing you to separate end-to-end planning from active building loops with structured progress tracking.

What safeguards are available for running coding agents in an isolated environment?▼

Safeguards for running coding agents include backpressure iteration control to limit loops, sandboxing options, and robust safe work-dir isolation to protect the host environment.

Can I use this workflow to coordinate Claude Code and Goose agents together?▼

Yes, you can coordinate multiple interactive coding agents including Claude Code and Goose by driving them through a unified exec with pty workflow for real TTY sessions.

Where are the updates and plans persisted during an automated agent building loop?▼

Updates and plans during an automated agent building loop are persisted in IMPLEMENTATION_PLAN.md, with supporting context stored in PROMPT.md and AGENTS.md files.

Related Skills

Explore More Software Engineering Skills→
tiankong0101-bytetiankong0101-byte

ralph-loop

Orchestrate interactive coding agent loops with TTY-safe execution and background monitoring.

Community
Advanced
rahulrajaramrahulrajaram

handoff

Commit code changes and update project documentation for AI session handoffs.

Community
Intermediate
dioerdendioerden

coding-agent

Execute coding tasks with Codex, Claude Code, OpenCode, and Pi agents.

Community
Advanced
KjdraganKjdragan
1

coding-agent

Control coding agents programmatically via PTY sessions across Codex, Claude Code, OpenCode, and Pi.

Community
Advanced