agentleFS
frankxaifrankxai/starlight-agent-skills/skills/education/coding-challenge-generator/SKILL.md

coding-challenge-generator

Skill1 starsChanged 4 months agoMITView on GitHub

Generate a graded set of coding challenges around a topic — progressive difficulty, each with statement, constraints, hidden tests, and solution. Use when building a problem set, a challenge ladder, interview prep, or a curriculum of exercises.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

Files
SKILL.md

Coding Challenge Generator

A topic + difficulty range → a graded ladder of auto-checkable challenges.

Purpose

Single exercises are easy; a coherent ladder that builds a skill is hard. This skill emits a set of challenges of rising difficulty around one topic, each fully specified and testable — ready to drop into a course, a bootcamp, or interview prep. (For a single richly-scaffolded space-themed lab, use ../../coding/cosmic-code-lab.)

When it fires

  • Keyword triggers: coding challenges, problem set, exercises, challenge ladder, interview prep
  • File triggers: challenges/**, problems/**
  • Command triggers: /coding-challenge-generator

Inputs

InputRequiredNotes
topicyesSkill/concept the set trains (e.g. "recursion").
languagenopython (default) or javascript.
countnoNumber of challenges, default 5.
rangenoDifficulty span, default intro→advanced.

Workflow

  1. Decompose the topic into a rising sequence of sub-skills. ✅ Check: each step depends on the prior.
  2. Per challenge: statement, constraints, input/output spec, hidden tests (happy + edge + error), and a reference solution that passes them.
  3. Grade the ladder — assign a difficulty label; verify monotonic increase.
  4. Self-verify — each solution passes its tests; each empty starter fails.

Output contract

{
  "topic": "string",
  "challenges": [{
    "title": "string",
    "difficulty": "intro | easy | medium | hard | advanced",
    "statement": "string",
    "starter_code": "string",
    "tests": "string",
    "solution": "string"
  }]
}

Tools & MCP

  • Tools: Write, Bash (run tests). MCP: none required.

Quality bar

  • Difficulty increases monotonically and is justified.
  • Each solution passes its tests; each starter fails them.
  • Statements are implementable from text alone.

Example

Input: topic "recursion", count 5 → five Python challenges from factorial to a tree traversal, each with hidden tests and a passing solution.


Built on SIP — Starlight Intelligence Protocol Substrate: starlightintelligence.org/protocol v1.1.0 Layers used: [file-contract, attestation, sovereignty] Vertical: starlight-agent-skills · portable capability layer

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.