Skip to content
Vibemetric

Free AI coding skill assessment

Get better at AI coding.

Learn what you’re doing wrong in Claude Code and Codex. Find what to improve. Become a better AI engineer.

Download free for Mac

No account required · Private by design · macOS 14+

Terminal
45
45/ 100Jul 6
  • Planning & Specs3
  • Context Engineering5
  • Memory & Knowledge5
  • Tool Integration6
  • Orchestration I: Task Mapping4
  • Orchestration II: Feedback Loops5
  • Testing & Evals5
  • Human Steering6
  • Security3
  • Efficiency & Cost3

Next: Write the pass/fail checks before the AI starts

Easy to start. Hard to master.

Using AI to code doesn’t mean you’re using it well.

Most developers can prompt Claude Code or Codex. Few know whether they follow good engineering practice while they do it.

  • Are you giving the AI enough context about your project?
  • Do you plan, and agree on what done looks like, before the AI writes code?
  • Do you test AI-generated code, or trust that it works?
  • Do you use skills, tools and agents, or explain the same things every time?
  • Do you spend more on AI models than each task needs?

How you learn with Vibemetric

  1. STEP 1

    Analyze your AI coding sessions

    Vibemetric reads the Claude Code and Codex sessions already on your Mac. No uploads, no questionnaires.

  2. STEP 2

    Find your strengths and weaknesses

    Get an overall score out of 100, and a score from 0 to 10 for each of 10 AI engineering skills.

  3. STEP 3

    Learn how to improve

    See the reason behind every score and the change to make first. Pro adds deep dives with recommendations for your own projects.

  4. STEP 4

    Track your progress

    Apply what you learned, scan again, and see which skills improved.

What you’ll learn

10 essential AI engineering skills.

Vibemetric scores each skill from your real sessions, so you know which one to work on next.

  1. 01

    Planning & Specs

    Do you and the AI agree on the outcome, how it’s checked, and the constraints before work starts?

  2. 02

    Context Engineering

    Does each project have a concise, well-structured CLAUDE.md or AGENTS.md that covers what AI must know, kept current?

  3. 03

    Memory & Knowledge

    Is AI memory kept accurate, so stale or mixed-up memories don’t cause mistakes?

  4. 04

    Tool Integration

    Does your recurring work run through tools and skills by default, and do you share the proven ones?

  5. 05

    Orchestration I: Task Mapping

    Is the work broken into the right pieces and wired together: order, inputs and outputs, and who does each?

  6. 06

    Orchestration II: Feedback Loops

    Do results decide the next step (fix, recheck, stop, escalate) and does the loop improve from experience?

  7. 07

    Testing & Evals

    Is checking a habit: tests that fit each project, real-flow checks of changes, and regular evals for AI features?

  8. 08

    Human Steering

    Do you examine the AI’s work and redirect it with clear reasons, then check the effect?

  9. 09

    Security

    Is the agent limited to what each task needs, and is what it builds checked for security problems?

  10. 10

    Efficiency & Cost

    Do models, effort, and usage match each task’s difficulty, adjusted from actual results?

Personal feedback, not generic tutorials

Stop guessing what you need to learn next.

  • Tutorials teach everyone the same thing. Vibemetric starts from your own sessions.
  • Best-practice lists don't know your weak spots. Your scores show which skill to work on first.
  • Courses end when the videos end. Scan again after each change and see what moved.

ILLUSTRATIVE FEEDBACK

Planning & Specs

4/10

WHAT WE FOUND

You often start coding tasks without saying how the result will be checked.

HOW TO IMPROVE

Before a feature, ask Claude Code for a short plan with clear success criteria and a checklist to verify them.

Before you write code, give me a short plan: what a user must be able to do, how we will check each point, and what must not change. Wait for my OK.

PRIVACY FIRST

Vibemetric never sees your transcripts.

Your sessions are read on your Mac and scored by your own Claude Code or Codex, under your own account. Here’s exactly what happens to them.

No Vibemetric server sees them

Scoring runs through your own Claude Code or Codex, exactly as when you use it yourself. The free app has no account, and Vibemetric’s server never receives your sessions. Pro only sends your API key to check your plan.

Read-only, all the way down

Vibemetric never writes to ~/.claude or ~/.codex. The Claude run it starts can only Read, Grep and Glob. No shell, no edits, no web.

Secrets are stripped first

Before anything is read, the digest removes API keys, tokens, webhook URLs and database connection strings it recognizes.

You decide what stays

Each result keeps its redacted digest on your Mac until you delete the result. Your CLI saves the scoring run in its history, but Vibemetric leaves it out of later scores, so it can’t skew them.

Your next level in AI coding starts with how you work today.

Free for Mac. Needs Claude Code or Codex installed and signed in. macOS 14+ · Apple Silicon and Intel · 5 MB download

Want more? Pro adds deep dives and recommendations for your own projects, for $9.99 a month.

Analyze my AI coding skills — free