claude-code

Automate coding and refactoring tasks via the Claude Code CLI.

4|Updated May 18, 2026
One-click install
npx skills add https://github.com/ZardLi1115/zedclaw --skill claude-code-zardli1115
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: claude-code
Source: https://github.com/ZardLi1115/zedclaw/tree/main/skills/autonomous-ai-agents/claude-code
Command: npx skills add https://github.com/ZardLi1115/zedclaw --skill claude-code-zardli1115

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

ZedClaw helps you delegate real coding work—refactors, fixes, and PR-sized changes—to Claude Code, without manually steering every edit and command.

Core Features & Use Cases

  • One-shot print mode for automation: Run non-interactive coding tasks that execute and return results.
  • Interactive PTY sessions for iterative development: Use tmux-backed sessions for multi-turn refactor → review → fix workflows.
  • Git workflow + PR-oriented execution: Generate changes, continue sessions, and support PR review loops using Claude Code features.
  • Safety/permission handling guidance: Provide deterministic handling for workspace trust and permission dialogs when using “dangerously-skip-permissions”.

Quick Start

Ask Claude Code to refactor a module and implement tests in one run by giving the instruction: Refactor the auth module and add unit tests, then verify the tests pass.

Frequently Asked Questions about claude-code

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

FAQPage Schema
How do I automate code refactoring and generate PR-ready changes?

You can automate code refactoring by delegating project-scoped modifications to the Claude Code CLI, which executes autonomous coding tasks and outputs PR-ready changes. Use one-shot print mode for scripted refactors and unit test generation.

Can I run interactive coding sessions with continuing context for multi-turn workflows?

Yes, interactive coding sessions with continuing context are supported via tmux-backed PTY orchestration. This allows multi-turn refactor, review, and fix workflows by maintaining session state for iterative development.

Do I need Claude Code installed and authenticated to use this agent orchestration?

Yes, a correctly configured Claude Code installation with proper authentication is required. The Skill delegates coding tasks to the Claude Code CLI, so valid setup is necessary to execute print-mode or tmux-based interactive commands.

How does this handle tool permissions and workspace trust during autonomous coding?

Tool permissions and workspace trust are managed through deterministic handling guidance, including options like dangerously-skip-permissions. This ensures controlled execution when running autonomous coding and refactoring tasks.

What is the best way to run non-interactive coding tasks for continuous integration?

The best way to run non-interactive coding tasks is using one-shot print mode. This executes scripted changes, verifies tests, and returns results without requiring manual steering or interactive terminal input.

How do I continue a previous coding session to review and fix generated changes?

You can continue a previous coding session using Claude Code's session continuation features. This supports PR review loops by maintaining context across multi-turn git workflows and iterative refinements.