skills.rest
tmtgroupbot
tmtgroupbotCOMMUNITY
Explore all skills by this author →
GitHub

coding-agent

Guides coding workflows through planning, execution, verification, and user-controlled preference storage.

Updated Mar 30, 2026
Search Tags:#planning#testing#software development#verification#code implementation#coding workflow#preference memory
One-click install
npx skills add https://github.com/tmtgroupbot/lptracker-deploy --skill coding-agent-tmtgroupbot
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: coding-agent
Source: https://github.com/tmtgroupbot/lptracker-deploy/tree/main/skills/coding-agent
Command: npx skills add https://github.com/tmtgroupbot/lptracker-deploy --skill coding-agent-tmtgroupbot

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Unstructured coding sessions often lead to untested code, oversized changes, and forgotten user preferences. This Skill provides a disciplined workflow for planning, executing, and verifying code changes while keeping the user in control of every step. ## Core Features & Use Cases - Structured Planning: Breaks multi-step implementation requests into independently testable steps with clear outputs and verification criteria. - Verification Guidance: Enforces test runs, screenshot reviews, and output checks before any code is delivered. - Preference Memory: Stores explicitly requested coding preferences in ~/code/memory.md so future sessions respect your style. - Use Case: When asked to build a login page, the Skill breaks the work into testable steps, tracks progress across multiple requests, suggests running tests after each function, and waits for your approval before proceeding. ## Quick Start Ask the agent to implement a feature and have it break the work into testable steps with verification after each one.

Frequently Asked Questions about coding-agent

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

FAQPage Schema
How do I break a large coding request into testable steps?▼

Define each step with a clear output and a test that verifies it, following the format: what is built, what exists after, and how to verify. Avoid vague steps like "implement the thing" and ensure each step is independently testable.

How does the coding workflow verify code before delivery?▼

Verification suggests running tests after each function, taking screenshots for UI changes, and running the full test suite before delivery. Screenshots are reviewed and fixed before being shown, never sent with known issues.

Does this skill execute code automatically?▼

No, it provides guidance only and never executes code autonomously. The user approves each step before execution begins, and sub-agent delegation requires an explicit user request.

Where are coding preferences stored and when?▼

Preferences are stored in ~/code/memory.md only when the user explicitly asks to save them, such as saying "remember I prefer TypeScript". One-off requests and project-specific requirements are never saved.

What are the limitations of this coding workflow skill?▼

It does not execute code, make network requests, or access files outside ~/code/ and the user's project. It is a guidance layer, so users wanting autonomous code generation pipelines need a different approach.

Related Skills

Explore More Software Engineering Skills→
arugo11arugo11
1

startproject

Analyze codebases with agent teams to generate project briefs and implementation plans.

Community
Advanced
saolalabsaolalab
37

git

Clone, branch, commit, and push code repositories via the git command line tool.

Official
Basic
travelinman1013travelinman1013

requesting-code-review

Automate pre-commit verification of staged code diffs with static scans, tests, and linting.

Community
Intermediate
RefoundAIRefoundAI
1.2k

vibe-coding

Guide non-technical teams to build runnable prototypes with AI coding tools.

Official
Intermediate