Skip to content
← Tools / Skills
Skill · .md packagebelbanas/10 trustMetrics-checked
Belbanas Agent Skills logo

Belbanas Agent Skills

Socratic coding coach and AI code teach-back skills for developers

Visit official site →

About

Two developer coaching skills built for personal use and shared as open source. socratic-coding-coach plays the role of a Socratic mentor, asking probing questions instead of handing over answers, while code-teach-back requires explaining AI-generated code back in plain terms before committing it. They are lightweight prompts, not tooling, so they work anywhere and are easy to adopt — but they also depend entirely on the model actually following the instructions.

What it can do

  • Socratic-style coding mentorship via socratic-coding-coach
  • Explain-before-commit code review via code-teach-back

Setup

  1. Install with the skills CLI:

npx skills add belbanas/agent-skills

or add the repository as a plugin marketplace in Claude Code and install the skill.

  1. Restart Claude Code so the new skill is discovered.

Prerequisites

Claude Code or another agent that reads SKILL.md files.

Use cases

learning to code with guided questioning instead of answersreviewing AI-generated code before committing itbuilding better habits around AI pair programming

Works with

Claude CodeAny SKILL.md-compatible agent

Trust score — 4/10

/10 trustMetrics-checked

Scored 4/10 (metrics tier, metrics-v1). Evidence: Star count not confirmed during verification; githubStars set to null rather than guessed.; Two prompt-only skills; quality depends entirely on model following instructions.; MIT license confirmed on GitHub license field.. License: MIT License, verified via GitHub license field..

How the trust score is built

Full methodology →
Adoption · stars/downloads, log-scaled vs peers
1.0/10
Maintenance · recency, releases, issue load
3.0/10
Openness · license + docs
8.0/10
Manual review · sanity read + red-flag check
4.0/10
  • · Star count not confirmed during verification; githubStars set to null rather than guessed.
  • · Two prompt-only skills; quality depends entirely on model following instructions.
  • · MIT license confirmed on GitHub license field.

Sources

Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.