skills.rest
annquin
annquinCOMMUNITY
Explore all skills by this author →
GitHub

coding-agent

Delegate coding tasks to AI agents in isolated background sessions.

Updated Apr 11, 2026
Search Tags:#coding#ai-agents#git-workflow#pr-review#pty#background-processing
One-click install
npx skills add https://github.com/annquin/revit-bim-workspace --skill coding-agent-annquin
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: coding-agent
Source: https://github.com/annquin/revit-bim-workspace/tree/main/skills/coding-agent
Command: npx skills add https://github.com/annquin/revit-bim-workspace --skill coding-agent-annquin

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill delegates coding tasks to AI agents via background sessions to accelerate software development, enabling teams to offload repetitive or complex coding work to automated helpers while maintaining control over the execution context.

Core Features & Use Cases

  • Background coding sessions with Codex, Claude Code, OpenCode, or Pi, including optional PTY-enabled terminals for interactive tasks.
  • PR reviews in isolated temporary directories, automated feature exploration, and incremental refactoring workflows.
  • Safe, isolated execution with explicit guidance on session lifecycle, monitoring, and result capture for reproducible outcomes.

Quick Start

Create a temporary git repository, start a PTY background coding session, and give the agent your task.

Frequently Asked Questions about coding-agent

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

FAQPage Schema
How do I delegate coding tasks to AI agents for automated PR reviews?▼

AI agents conduct PR reviews in isolated temporary directories using PTY-enabled background sessions. This approach safely captures monitoring data and execution results for iterative code exploration and incremental refactoring workflows.

Can I use Claude Code or Codex for background coding sessions?▼

Background coding sessions support Codex, Claude Code, OpenCode, and Pi with per-agent tooling configuration. These agents operate within PTY-enabled terminals to execute interactive coding tasks in isolated temporary workspaces.

What is the best way to automate feature development with AI coding agents?▼

Automating feature development requires delegating tasks to AI agents in isolated temporary workspaces with explicit session lifecycle monitoring. Background sessions enforce reproducible execution rules to maintain controlled development outcomes.

Do I need PTY-enabled terminals to run AI coding agents in the background?▼

PTY-enabled background sessions are required for interactive AI coding tasks. The skill uses PTY terminals to support per-agent tooling, monitor session lifecycles, and capture reproducible execution results in isolated workspaces.

How does isolated execution work for AI agents performing code refactoring?▼

Isolated execution confines refactoring tasks to temporary git repositories within background sessions. This safe isolation ensures the AI agent operates in a monitored, reproducible environment, preventing uncontrolled changes to the main codebase.

When should I not use background AI agents for software development?▼

Avoid background AI agents when you lack safe isolation for temporary workspaces or cannot monitor session lifecycles. Reproducible execution requires controlled environments, making it unsuitable for contexts without explicit per-agent tooling support.

Related Skills

Explore More Software Engineering Skills→
rongarederongarede
4

collaborating-with-codex

Delegate coding tasks to Codex CLI via a Python bridge with JSON outputs.

Community
Intermediate
keugenekkeugenek
12

shark-status

Check Shark background job status, completed results, and recent logs.

Community
Basic
pertain99pertain99
1

coding-agent

Delegate coding tasks to Codex, Claude Code, Pi, and OpenCode agents.

Community
Advanced
coltonbattscoltonbatts

coding-agent

Execute AI coding agents via bash commands for code generation and review.

Community
Advanced