skills.rest
jbreel77888
jbreel77888COMMUNITY
Explore all skills by this author →
GitHub

claude-code

Delegate coding, analysis, and refactoring tasks to Claude Code CLI.

Updated Aug 27, 2026
Search Tags:#code review#parallel execution#task delegation#claude-code#session resume#sub-agent#permission modes
One-click install
npx skills add https://github.com/jbreel77888/Agent-AiNorx --skill claude-code-jbreel77888
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: claude-code
Source: https://github.com/jbreel77888/Agent-AiNorx/tree/main/.claude/skills/claude-cli
Command: npx skills add https://github.com/jbreel77888/Agent-AiNorx --skill claude-code-jbreel77888

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill eliminates the friction of manually switching context to run coding, analysis, or refactoring tasks in Claude Code, while ensuring you retain full control and verification over all sub-agent outputs to avoid unreviewed changes or errors.

Core Features & Use Cases

  • Headless Sub-Agent Delegation: Run Claude Code in non-interactive mode to handle coding tasks without manual input, freeing you to focus on higher-level planning.
  • Flexible Permission Modes: Choose from least-privilege access levels (default, plan, acceptEdits, bypassPermissions) to match the risk level of each task, from read-only reviews to full implementation.
  • Supervision & Verification Workflow: Built-in guidance for capturing sub-agent outputs, independently verifying changes via git diff and test runs, and iterating on corrections via session resume.
  • Use Case: You need a second opinion on a complex refactor; use this Skill to delegate the task to Claude Code in plan mode, capture its proposed changes, and review them before applying.

Quick Start

Use the claude-code skill to delegate a code review of your current uncommitted changes to Claude Code in plan mode and capture the output for your review.

Frequently Asked Questions about claude-code

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

FAQPage Schema
How do I delegate coding tasks to Claude Code as a non-interactive sub-agent?▼

You can delegate coding tasks to Claude Code as a non-interactive sub-agent by running it in headless mode from a host environment. This allows the CLI to handle analysis or refactoring autonomously while you retain full supervision and verification control over the outputs.

Can I run parallel execution of independent coding tasks using Claude Code?▼

Yes, parallel execution of independent coding tasks is supported by delegating them to Claude Code sub-agents. This enables you to handle multiple refactoring or analysis tasks simultaneously without manual context switching.

What permission modes are available for task delegation in Claude Code?▼

Available permission modes include default, plan, acceptEdits, and bypassPermissions. These allow you to enforce least-privilege access levels, matching the risk profile of delegated tasks from read-only plan reviews to full implementation.

How do I verify sub-agent outputs before applying unreviewed changes?▼

You verify sub-agent outputs by independently checking changes via git diff and running tests before acceptance. This supervision workflow prevents unreviewed changes and errors from being applied to your codebase.

Can I resume a session to iterate on multi-turn task refinement with Claude Code?▼

Yes, you can resume a session to iterate on multi-turn task refinement with Claude Code. This allows you to correct sub-agent outputs and continue refining complex coding tasks without losing previous context.

What is the best way to get an adversarial code review from Claude Code?▼

The best way to get an adversarial code review is delegating the task to Claude Code in plan mode. This captures proposed changes for your independent review without applying edits, ensuring a safe second opinion on complex refactors.

Related Skills

Explore More Software Engineering Skills→
neubigneubig
3

sub-agent-delegation

Delegate parallel tasks to sub-agents via DelegateTool or OpenHands Cloud API.

Community
Advanced
DataDogDataDog
648

run-tests

Identify changed files and run affected test suites via scripts/run-tests.

Official
Advanced
bobmatnycbobmatnyc
146

Requesting Code Review

Dispatch a code-reviewer subagent to review changes against plans and requirements.

Community
Basic
jeannotlemaire-uxjeannotlemaire-ux

design-taste-frontend-v1

Preserve v1 taste-skill behavior with design and engineering directives.

Community
Advanced