Skip to content
Vibemetric

forClaude CodeandCodex

The cheat code for AI coding

Faster shipping, safer code, smaller token bills. Vibemetric finds what’s holding you back and writes the fix.

  • Token usage. Prompt: “change the buy button label to “Buy now” on both mobile and desktop, and use opus for this one so it’s done properly”. Vibemetric: Expensive model for a light task (Efficiency & Cost): Changing a label is a small-model job. Opus costs several times more for the same result.
  • Security. Prompt: “ok here's the live key sk_live_••••••4242, just put it straight in config.ts so it works on vercel too, then commit and push to main”. Vibemetric: Live secret pasted into chat (Security): It now sits in your session history. Rotate it, and let the app read it from the environment. A secret headed for git (Security): Keep keys in .env.local, listed in .gitignore, and check the diff before any push.
  • Skills. Prompt: “now open the site in a browser, click through the whole signup flow on desktop and mobile, and tell me if anything breaks or looks off”. Vibemetric: There’s a skill for this (Tool Integration): The webapp-testing skill drives a real browser and reports what broke. Install it once, then just ask.
  • Context. Prompt: “follow our usual conventions for this one, make it match the look of the other settings screens, and keep it simple so it ships today”. Vibemetric: “Usual” isn’t written anywhere (Context Engineering): Put the conventions in CLAUDE.md, so every session starts with them. Name the files to copy (Context Engineering): Add @app/settings/Profile.tsx to the prompt, so it matches the right screen, not a guess.

Built by engineers from

  • LinkedIn
  • Intel
  • eBay

The hidden cost of building with AI.

  • Cost

    30×

    Same task. Unpredictable spend.

    Token costs varied by up to 30× between identical agent runs.

    Stanford Digital Economy Lab2026

  • Code quality

    +81%

    More duplicated code.

    Duplication rose as meaningful refactoring fell away.

    GitClear623M code changes · 2026

  • Security

    1 in 2

    Code samples failed security tests.

    AI-generated code still leaves critical vulnerabilities behind.

    VeracodeGenAI Code Security Report · 2026

  • Productivity

    19%

    Slower than coding by hand.

    Experienced developers took longer, yet felt 20% faster.

    METRRandomized controlled trial · 2025

Introducing Vibemetric

See what to fix, and why.

One click, and it’s fixed.

Each card comes with a fix prompt. Click Run in Claude Code, and it makes the change for you.

See how you’re improving.

Run a full score anytime: all ten categories, weakest first, each with what to fix.

Vibemetric

Today 9:12am

85/ 100

↑ 7 since Nov 2

Focus on these next

  • Make the agent run the tests after every change

    What we saw
    Tests ran after only 8 of 31 code changes this week.
    Why it matters
    Two tasks were reported done and broke the next day.

    Add this to your context file or prompt

    After every code change, run npm test and show me the result. Don’t say a task is done until the tests pass.

    Copy

About this score

How Vibemetric works

  1. Vibemetric scans your AI coding sessions

    It reads the Claude Code and Codex sessions already on your Mac. Nothing to upload, nothing to fill in.

  2. Finds what’s holding you back

    It spots the habits that cost you time, tokens and rework, and tells you why each one matters and what to change.

  3. Writes the fix for you

    Each problem comes with a prompt made for your project. Run it straight from the app, or copy and paste it into Claude Code or Codex.

PRIVACY BY DEFAULT

The scanning happens on your Mac. Scoring runs through your own Claude Code or Codex, under your account. Vibemetric’s server never sees your sessions.

How privacy works

Ten categories. Graded from your real sessions.

  • Planning & Specs

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

  • Context Engineering

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

  • Memory & Knowledge

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

  • Tool Integration

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

  • Orchestration I: Task Mapping

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

  • Orchestration II: Feedback Loops

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

  • Testing & Evals

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

  • Human Steering

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

  • Security

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

  • Efficiency & Cost

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

  • Your score

    65 / 100

    You give AI real tools it keeps reusing, and it checks live behavior on its own. But you rarely say what “done” means up front, so design tasks end in rounds of taste corrections.

Swipe for all 10

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. Vibemetric’s server never receives your sessions or transcripts. It gets only each score’s numbers, to keep your history and pick the cards pinned in your Feed.

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.

Pricing

Your score is free. Pro helps you raise it.

Free

$0forever

The full score, on your Mac.

  • AI Native Score out of 100
  • All 10 dimensions, each with the case that earned it
  • Recommendations for every dimension
  • Score history with what changed since last time
  • Menu bar score, Markdown and image export

Pro

Grand opening

Regular price $19.99$9.99per month

Grand-opening price for new members50% off

7-day free trial

Everything in Free, plus tools and progress from our server.

  • Specific tool picks for each dimension, with setup steps
  • A curated tool catalog, updated every week
  • How you compare with other Vibemetric users
  • Progress over time, and whether each change worked
  • Team view: see your team’s scores, gaps, and trends

Then $9.99 per month. Cancel anytime.

Questions

What does it cost?

The app and the full score are free. Pro is $9.99 a month or $99 a year, with a 7-day free trial, and adds tool picks, comparisons, progress tracking and a team view. Scoring itself runs on your existing Claude plan through the claude command you already have installed, like any other Claude Code session.

Why does scoring take five minutes?

The rubric only counts connected evidence: a rule saved in one session and used in a later one, or a failed check that was fixed and rerun. Finding those takes reading, not counting.

Is the score stable?

Scores come from an AI reviewer, so they can move a point or two between runs. Reasons always cite the sessions they come from, so you can check them.

Why does macOS ask before it opens the app?

macOS asks once about any app downloaded from the internet. Vibemetric is signed and notarized by Apple, so click Open. Updates install automatically after that.

Contact

Questions or ideas? Write to us.

Feedback, bug reports and team questions all welcome. We reply by email.

Get your score.

Open the .dmg, drag Vibemetric into Applications, and open it. Needs Claude Code or Codex installed and signed in.

macOS 14+ · Apple Silicon and Intel · 12 MB download

Download for Mac