coding-agent

Orchestrate autonomous coding agents for background development tasks.

Updated Jun 18, 2026
One-click install
npx skills add https://github.com/wangqianCAI/OBI --skill coding-agent-wangqiancai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: coding-agent
Source: https://github.com/wangqianCAI/OBI/tree/main/skills/coding-agent
Command: npx skills add https://github.com/wangqianCAI/OBI --skill coding-agent-wangqiancai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill solves the challenge of managing long-running, autonomous coding tasks by providing a structured, background-process-oriented workflow that ensures completion, notification, and proper environment isolation.

Core Features & Use Cases

  • Background Execution: Offloads heavy coding tasks like refactoring or feature building to background processes, keeping your chat interface responsive.
  • Multi-Agent Support: Seamlessly integrates with Claude Code, Codex, OpenCode, and Pi agents.
  • Reliable Completion: Enforces a mandatory notification route pattern, ensuring you are alerted immediately when a task finishes or fails.
  • Use Case: Use this to trigger a background refactoring of a large module or to run a batch of PR reviews across multiple git worktrees without blocking your current session.

Quick Start

Use the coding-agent skill to start a background task for building a new feature by providing the task description and the required notification route.

Frequently Asked Questions about coding-agent

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

FAQPage Schema
How do I run background coding agents for feature implementation without blocking my chat session?

Run background coding agents by spawning isolated shell processes with specific CLI flags. This offloads heavy coding tasks like feature implementation to background processes, keeping your chat interface responsive while maintaining persistent session monitoring.

Can I trigger automated pull request reviews across multiple git worktrees simultaneously?

Yes, automated pull request reviews can be run in batches across multiple git worktrees. The skill orchestrates autonomous coding agents in isolated background processes, allowing you to execute multiple reviews concurrently without blocking your current session.

Does this autonomous coding agent support Claude Code and Codex models?

This autonomous coding agent supports multi-agent integration with Claude Code, Codex, OpenCode, and Pi agents. It operates by spawning isolated shell processes with specific CLI flags tailored for each coding model to execute background development tasks.

How do autonomous coding agents notify task completion status during long-running refactoring?

Autonomous coding agents enforce a mandatory notification routing protocol to communicate task completion status back to the user. This ensures you are alerted immediately when a task finishes or fails, without relying on system heartbeats.

What is the best way to orchestrate code refactoring tasks in isolated environments?

The best way to orchestrate code refactoring is to offload the task to a background coding agent. The skill spawns isolated shell processes for environment isolation, ensuring persistent session monitoring and immediate notification upon task completion or failure.

Are there limitations to running long-running coding tasks without system heartbeats?

Long-running coding tasks require strict adherence to notification routing protocols because system heartbeats are not supported. You must configure the mandatory notification route pattern to ensure task completion or failure status is communicated back to the user.