skills.rest
lofty-thoughts
lofty-thoughtsOFFICIAL
Explore all skills by this author →
WebsiteGitHub

coding-agent

Delegate coding tasks to interactive agents via a bash-based workflow.

Updated Feb 21, 2026
Search Tags:#automation#bash#agents#interactive#coding#pty
One-click install
npx skills add https://github.com/lofty-thoughts/botdaddy --skill coding-agent-lofty-thoughts
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: coding-agent
Source: https://github.com/lofty-thoughts/botdaddy/tree/main/seed/skills/coding-agent
Command: npx skills add https://github.com/lofty-thoughts/botdaddy --skill coding-agent-lofty-thoughts

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Delegates intensive coding work to interactive agents, reducing manual time spent on repetitive or exploratory programming tasks.

Core Features & Use Cases

  • Background PTY-enabled coding sessions: Run coding agents in the target workspace with a persistent interactive terminal.
  • Use cases: Build features, review PRs, refactor large codebases, and perform iterative exploration that explores project context.
  • Use Case Example: Imagine you need to implement a new feature across multiple files; spawn Codex or Claude Code in a focused directory to draft changes, review diffs, and push updates autonomously.

Quick Start

Launch a coding agent in your project workspace to begin an interactive session.

Frequently Asked Questions about coding-agent

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

FAQPage Schema
How do I automate coding tasks using interactive agents in bash?▼

To automate coding tasks using interactive agents, this Skill delegates work like feature building and refactoring to a bash-based workflow. It requires a PTY-enabled bash tool and an agent pool such as Codex or Claude Code to execute autonomously.

Can I run background coding sessions for refactoring large codebases?▼

Yes, you can run background coding sessions for refactoring large codebases. The Skill spawns interactive agents in a persistent PTY-enabled terminal, allowing them to explore project context and apply changes across multiple files autonomously.

Do I need a PTY-enabled bash tool to delegate PR reviews?▼

Yes, a PTY-enabled bash tool is required to delegate PR reviews. The Skill relies on this persistent interactive terminal environment to run agent pools like Codex, Claude Code, OpenCode, or pi within your target workspace.

What is the best way to perform exploratory coding across multiple project files?▼

The best way to perform exploratory coding across multiple project files is by spawning a coding agent in your focused workspace directory. The agent uses the interactive terminal to explore project context, draft changes, and review diffs iteratively.

Which interactive agent pools work with this bash-based coding workflow?▼

The interactive agent pools that work with this bash-based coding workflow include Codex, Claude Code, OpenCode, and pi. You must configure one of these supported agent pools to execute the delegated coding tasks.

Why does exploratory coding require a persistent interactive terminal?▼

Exploratory coding requires a persistent interactive terminal because the agents need an active PTY session to maintain file-system context. This allows the background agents to execute bash commands, read project files, and apply iterative code changes continuously.

Related Skills

Explore More Software Engineering Skills→
DharambirAgrawalDharambirAgrawal

coding-agent

Run Codex, Claude Code, OpenCode, or Pi agents via PTY-enabled Bash sessions.

Community
Intermediate
2460124601

ship

Merge base branch, run tests, bump VERSION, update CHANGELOG, and open a pull request.

Community
Advanced
bruinalphainvestmentbruinalphainvestment

openspec-explore

Explore codebases and create OpenSpec artifacts for change proposals.

Official
Advanced
SignmanalSignmanal
1

python-debugpy

Debug Python code across local, remote, post-mortem, and test scenarios.

Community
Intermediate