Build From Scratch: 25 classic programs to build with an AI agent as your tutor, not your typist
by PaperChaseWebb
A free guide to the 25 classic build-your-own projects (shell, scheduler, database, file system, version control, web server, compiler and more), sorted into three tiers with a finishable v1, what each teaches, the first test and what to leave out. Includes a tutor rules block that keeps Grok Bot, Claude Code, Codex or Hermes explaining and reviewing instead of writing your code, five approval steps, six prompts, a five-milestone shell starter and a tested milestone check script.
- Price
- Free
- Difficulty
- Intermediate
- Setup
- 20 min
- Format
- .md file
build-from-scratch-25-builds-ai-tutor-free.md · 22 KB · direct download, no signup
What you end up with
You pick one build, scope a v1 you can finish, and work through it in milestones with an agent that teaches and reviews instead of typing, starting with a small shell whose every milestone is checked by a script.
What you get
- 25 builds in three tiers, each with a finishable v1, what it teaches, the first test and what to leave out
- A tutor rules block for Grok Bot, Claude Code, Codex or Hermes
- Five approval steps so the agent asks before scope, tests, outside commands and full answers
- A milestone format to copy
- A small shell starter in five milestones
- shell_check.py: 14 checks across four milestones, tested against a reference and planted faults
- Six prompts: pick, scope, plan, teach, review, write up
- Harness notes and a failure table
- Credits to the lists and courses we looked at
Who it is for
Developers who use AI coding agents every day and want to understand the layers underneath them, without letting the agent do the learning for them.
Works with
Solid chips: the agent is named in this listing's own tools list. Dashed chip: Plain Markdown, so it also pastes into other agents such as Grok Bot, Hermes, OpenClaw or Muse. We list those as compatible by format only, not as tested.
Tools needed
Cost to run: Free. No signup.
Time to result: About 20 minutes to pick and scope a build; the shell starter runs over a few sessions at your pace
How it works
Pick one build
Start in tier one, or run the pick prompt.
Expected: One build and a backup.
Load the tutor rules
Paste the rules block or save it as the project rules file.
Expected: An agent that explains and reviews, and stops at rule 1.
Scope v1
One page, five bullets or fewer, with a named v2 list.
Expected: SCOPE.md you approved.
Work in milestones
One new idea per milestone, each ending with a passing test. For the shell, run shell_check.py with --upto.
Expected: A program that runs after every milestone.
Write it up
What broke, and what the real tool was doing for you.
Expected: A one-page write-up in your own words.
Sample prompts, free to use
I am starting milestone [n]. Explain the one new idea in plain words, the system calls or data structures involved and the most common mistake people make with it. Point me to the real documentation. Then stop. Do not write my code.
All six prompts and the tutor rules block are in sections 2 and 10 of the file.
Free preview of the file
# Build From Scratch: 25 classic programs to build with an AI agent as your tutor, not your typist A free guide by PaperChaseWebb for developers who use AI coding agents every day and want to understand the layers underneath them. The 25 classic "build your own" projects, sorted into three tiers, each with a v1 small enough to finish, what it teaches, the first test to write and what to leave out. Plus a tutor rules block that keeps Grok Bot, Claude Code, Codex or Hermes explaining and reviewing instead of writing the whole thing for you, six prompts, a five-milestone starter build (a small shell) and shell_check.py, a test script that checks each milestone for you. Plain .md. No signup, the whole thing is in this file. Most "build your own X" lists are a wall of links. You pick something huge, follow a tutorial that hands you the code, and two weeks later you have a folder you copied and cannot explain. With an agent in the loop it gets worse: ask it to "build a database" and it will, in one go, and you learn nothing. This guide fixes the two things that decide whether a from-scratch build teaches you anything: a v1 you can actually finish, and an agent that is only allowed to teach. ## Contents 1. How to use this 2. The tutor rules block (paste first) 3. Approval steps 4. Tier one: build first 5. Tier two: strong second builds 6. Tier three: mastery builds 7. The milestone format 8. Starter build: a small shell in five milestones 9. shell_check.py, the milestone test sc
Worth knowing
- A learning guide, not a course with grading. The tests check behavior, not code quality.
- shell_check.py was tested on Linux against our own reference shell and planted faults on 2026-10-27; the reference is not included so you build your own.
- Tier two container work and all tier three builds belong in a VM or container, not your main machine.
Questions
Why not just let the agent build it?
You can, and you will get a working program and learn very little. This guide is for the other goal: understanding the layer. The rules block keeps the agent to explaining, hinting and reviewing.
Which language should I use?
Any language with process, file and socket calls. C, Rust, Go and Python all work for the shell starter.
Does shell_check.py touch my files?
No. Each check runs your shell in a fresh temporary folder that is deleted afterwards.
What comes next?
If you want agents to write production code well, see Vibe Coding with AI Agents ($39). If you run a team and want agents on real workflows, start with the free Agent Readiness Self-Check.
Details
Version 1.0 · Updated Oct 27, 2026 · Markdown version of this page
License: Free for personal and commercial use, including teaching. Do not resell or republish this file. shell_check.py may be copied into your projects. Download terms
Related
Vibe Coding with AI Agents: method, prompts, skills and checklists
A Markdown playbook for building real software with Claude Code, Codex or Claude: a seven-step method, project file templates, 18 prompts, five agent skills, release and hardware checklists, cost notes and 24 project ideas.
Spec-First Build Docs: PRD, design system, architecture note, agents guide and readiness check
A Markdown pack of the four documents to write before an agent builds anything, sized to the stakes: a PRD an agent can build from without inventing scope, the smallest design system that holds, a one-page architecture note with reversal costs, an agents guide with a project tree for Claude Code, Codex and Cursor, a weekly staleness pass and a nine-area readiness check.
AGENTS.md, CLAUDE.md and Cursor rules starter files
Six fill-in instruction files for coding agents (base, Next.js plus Supabase, FastAPI, browser game, CLAUDE.md pointer, scoped Cursor rules) and 8 prompts to write, trim and test your own.
Revive Old Tech with Coding Agents: drivers, firmware, crash patches and device bridges
A Markdown method for making abandoned hardware and old software work again with a coding agent: a REVIVAL.md brief, six revival lanes, ways to let the agent see the target, a closed build and read loop, a bug-to-check ledger, do-no-damage rules, a rules block and 14 prompts.
Agent Readiness Self-Check: score one workflow before you hand it to an AI agent
A free self-check for agencies and small businesses. Score one workflow across eight areas (workflow mapped, data access, approval gates, owner per agent, handoffs, cost tracking, security and vetting, measurement) with an evidence rule and three gate rules, read your band and what to do next, plus an interview prompt for Grok Bot, Claude Code, Codex or Hermes, a score sheet and a worked example.