coding-agent

Plan, implement, and verify coding tasks with user approval.

2|Updated Mar 31, 2026
One-click install
npx skills add https://github.com/marktantongco/promptc-os --skill coding-agent-marktantongco
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: coding-agent
Source: https://github.com/marktantongco/promptc-os/tree/main/skills/coding-agent
Command: npx skills add https://github.com/marktantongco/promptc-os --skill coding-agent-marktantongco

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Coding teams and individuals need a structured, non-autonomous workflow to plan, implement, and verify software tasks while keeping user control and local memory intact. This guide provides a repeatable framework for breaking down coding requests, guiding execution, and validating outcomes without auto-execution.

Core Features & Use Cases

  • Planning templates and references (planning.md) to decompose complex requests into testable steps.
  • Execution guidance (execution.md) that progresses only with user approval and keeps progress visible.
  • Verification readiness (verification.md) with checks for outputs, tests, and UI validations, updating the user as work proceeds.
  • Local memory management (memory-template.md) for storing user preferences when requested and maintaining privacy.

Quick Start

Describe your coding task and let the agent generate a plan, implement steps, and verify results with user approval.

Frequently Asked Questions about coding-agent

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

FAQPage Schema
How do I plan and verify coding tasks step by step without autonomous execution?

A structured coding workflow coordinates planning, execution, and verification to prevent autonomous actions. It breaks down software tasks into testable steps, guides implementation only with user approval, and validates outputs through formal checks to ensure correctness.

What is the best way to break down complex software requests into testable steps?

The best way to break down complex software requests is using planning templates that decompose tasks into testable steps. This structured approach ensures execution progresses visibly and only with user approval, maintaining clear oversight throughout the coding workflow.

How does a non-autonomous coding workflow handle local memory and user preferences?

A non-autonomous coding workflow handles local memory by storing user preferences only when explicitly requested and maintaining privacy. It reads user memory during planning and execution to guide steps while preventing unauthorized autonomous actions within local workspaces.

Can I use a stepwise coding workflow to verify outputs and UI validations locally?

Yes, you can use a stepwise coding workflow to verify outputs, tests, and UI validations locally. It includes verification readiness checks that update you as work proceeds, ensuring software tasks are validated within your local workspace without auto-execution.

Does a structured coding workflow require specific dependencies to implement and verify software?

No, a structured coding workflow does not require specific dependencies to implement and verify software. It operates independently using planning, execution, and verification references to guide stepwise implementation and validate outcomes within local workspaces.