skills.rest
Wide-Moat
Wide-MoatOFFICIAL
Explore all skills by this author →
GitHub

sub-agent

Delegate complex multi-file code tasks to an autonomous sub-agent session.

112|26|Updated Jan 25, 2026
Search Tags:#code review#code refactoring#autonomous coding#model discovery#session resume#sub-agent#test-fix cycle
One-click install
npx skills add https://github.com/Wide-Moat/open-computer-use --skill sub-agent-wide-moat
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: sub-agent
Source: https://github.com/Wide-Moat/open-computer-use/tree/main/skills/public/sub-agent
Command: npx skills add https://github.com/Wide-Moat/open-computer-use --skill sub-agent-wide-moat

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

The sub-agent skill helps complete large, iterative code tasks that typically require many tool calls and back-and-forth fixes until tests and requirements finally pass.

Core Features & Use Cases

It spawns a separate autonomous sub-agent session for deep development work, and it enforces a strict task structure (ROLE/DIRECTIVE/CONSTRAINTS/PROCESS/OUTPUT) to reduce thrashing and ensure verifiable outcomes; it is intended for complex CODE tasks such as multi-file refactors with test loops, code review with automated fixes, and iterative test-fix cycles that exceed 10 tool calls.

Quick Start

Ask your AI to run the sub-agent skill with a task that includes ROLE, DIRECTIVE, CONSTRAINTS, PROCESS, and OUTPUT, and make sure the task includes a concrete verification command (for example, the test runner) and a clear stop condition.

Frequently Asked Questions about sub-agent

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

FAQPage Schema
How do I automate complex multi-file refactoring with iterative test-fix cycles?▼

You can automate complex multi-file refactoring by delegating the work to an autonomous sub-agent session that handles iterative test-fix loops and enforces a structured task template to ensure verifiable outcomes.

What is the best way to run multi-pass code reviews with automated fixes?▼

Multi-pass code reviews with automated fixes are best handled by spawning a separate autonomous sub-agent session that processes constraints and executes repeated test-run analysis cycles until requirements pass.

How do I structure tasks for an autonomous coding agent to prevent thrashing?▼

To prevent thrashing, structure autonomous coding agent tasks using a strict ROLE, DIRECTIVE, CONSTRAINTS, PROCESS, and OUTPUT template that includes a concrete verification command and a clear stop condition.

When do I need to use an autonomous sub-agent for code refactoring?▼

You need an autonomous sub-agent for code refactoring when your task involves large multi-file changes, multi-pass code reviews, or iterative test-fix cycles that require 10 or more tool calls.

Can I resume an autonomous coding session after hitting max_turns limits?▼

Yes, you can resume an autonomous coding session after hitting max_turns limits because the workflow includes session resume support to continue long-running iterative code tasks.

What limitations should I expect when delegating iterative code tasks to a sub-agent?▼

When delegating iterative code tasks to a sub-agent, expect max_turns limits on autonomous sessions and the requirement to perform pre-run model discovery to satisfy necessary guardrails before execution.

Related Skills

Explore More Software Engineering Skills→
Moliboy5000Moliboy5000

sciomc

Orchestrate parallel scientist agents for research decomposition, execution, verification, and synthesis.

Community
Advanced
etylsarinetylsarin
59

session-checkpoints

Save and restore session state in .opencastle/SESSION-CHECKPOINT.md files.

Community
Intermediate
zeronx798zeronx798

requesting-code-review

Scan staged git changes for security issues and quality regressions.

Community
Advanced
granietgraniet
247

claude-code

Delegate coding tasks to the Claude Code CLI for automation.

Community
Advanced