autonomous-loops

Decompose RFCs into a dependency DAG and coordinate multi-agent implementation loops.

2|2|Updated Mar 12, 2026
One-click install
npx skills add https://github.com/zhmxiaowo/opencode-simple --skill autonomous-loops-zhmxiaowo
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: autonomous-loops
Source: https://github.com/zhmxiaowo/opencode-simple/tree/main/.opencode/skills/autonomous-loops
Command: npx skills add https://github.com/zhmxiaowo/opencode-simple --skill autonomous-loops-zhmxiaowo

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Orchestrating complex Claude Code workflows across multiple agents, this skill coordinates actions, manages dependencies, and bridges context across iterations to enable autonomous software development.

Core Features & Use Cases

  • RFC-driven decomposition of large specifications into a dependency DAG.
  • Layered, context-isolated stages (research, plan, implement, test, review) executed by separate agents.
  • Parallel landings with a robust merge queue and eviction-context recovery.
  • Suitable for large RFCs, multi-unit features, and long-running product initiatives.

Quick Start

Provide an RFC document and run the orchestrator to decompose it into a dependency DAG and start the staged agent pipelines.

Frequently Asked Questions about autonomous-loops

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

FAQPage Schema
How do I decompose a large RFC into parallel work units for autonomous agents?

Decomposing an RFC involves breaking the specification into a dependency DAG, coordinating multi-agent loops to implement isolated stages like research, planning, and testing for parallel execution.

What is the best way to manage merge conflicts in multi-agent Claude Code workflows?

Managing merge conflicts in multi-agent workflows requires a robust merge queue and eviction-context recovery, ensuring parallel landings are tracked and merged without losing iterative context.

How do I orchestrate staged pipelines for complex software features?

Orchestrating staged pipelines uses context-isolated stages executed by separate agents, bridging context across iterations to enable autonomous software development for large features.

Do I need a dependency DAG to schedule autonomous Claude Code loops?

A dependency DAG is required to schedule and track parallel work units, allowing the orchestrator to map RFC requirements into tiered complexity guidance and per-stage isolation.

When should I use multi-agent orchestration for software development?

Use multi-agent orchestration for large RFCs, multi-unit features, and long-running product initiatives where parallel work units must be scheduled, tracked, and merged with conflict management.

Can I run research and implementation agents in isolation for the same feature?

Running research and implementation agents in isolation is supported through layered, context-isolated stages, ensuring separate agents handle distinct phases without cross-contamination.