You're using Claude Code.
But are you using it well?
prompt1 analyzes your habits across 10 proficiency categories and shows exactly where to improve — from prompt quality to context management to git discipline.
Managing a team? See team plans
Free foreverNo credit card2-min install
The proficiency gap
Where most developers start.
Where prompt1 takes you.
CLAUDE.md is a file that tells Claude Code about your project — coding standards, architecture, and preferences. Plan mode is a Claude Code feature that thinks through an approach before writing code.
Using AI as autocomplete
Gets some value, but leaves most of it on the table.
- –No CLAUDE.md — AI has zero project context
- –Short prompts without structure or intent
- –Skips plan mode — jumps straight to code
- –Uses more tokens than needed
Getting dramatically more from the same tool
Knows exactly how to get the most from Claude Code.
- ✓Rich CLAUDE.md — AI understands the codebase
- ✓Structured prompts with clear context
- ✓Plans first, then builds systematically
- ✓Efficient token usage — right context, right time
Small habit changes compound fast. Each coaching recommendation moves you closer.
Getting started
Two minutes to install
First coaching insights appear after about 1 hour of normal Claude Code usage.
Install the plugin
One command. Works with any Claude Code CLI setup on macOS, Linux, or WSL.
Want to inspect first? The plugin is plain shell scripts — read the source before running.
Use Claude Code normally
The plugin captures usage patterns in the background — prompt structure, tool usage, workflow choices. No changes to your workflow. Zero performance impact.
Get coached to improve
Your dashboard shows a proficiency score across 10 categories with specific, actionable recommendations.
Example coaching recommendation
Your CLAUDE.md is missing project context
Your CLAUDE.md file doesn't describe your tech stack or coding conventions. Adding 3–5 lines about your project architecture helps Claude Code give more accurate, context-aware responses.
The coaching dashboard
See where you stand. Get better every day.
Whether you started using Claude Code yesterday or six months ago, prompt1 shows you exactly what to improve — and how.
New to Claude Code?
Developers just getting started often improve the fastest — there's so much to discover. prompt1 shows you what to try next.
Each category is scored 0–100. Your overall proficiency rating (0–2000) combines category scores with your historical coding patterns.
10-Category Dashboard
Personalized Coaching
LLM-powered analysis of your Claude Code habits. Each category shows your score, what you're doing well, and specific recommendations to improve.
- ✓Instruction Design — how well your CLAUDE.md sets up AI context
- ✓Prompt Engineering — structure, clarity, and intent of your prompts
- ✓Context & Planning — plan mode usage, session management
- ✓Git, Safety, Verification — commit discipline, security practices
Profile & Rating
Shareable Developer Profile
A proficiency profile backed by real coding behavior — not self-reported skills. Share it when you're ready.
- ✓Overall rating: 0–2000 (Bronze → Diamond)
- ✓Public profile — private by default, share when ready
- ✓Global leaderboard ranking
- ✓Score improves as you build better habits
For engineering teams
Your team has the same gap. Multiplied.
You bought Claude Code licenses for your team. Are they all getting value? prompt1 gives you aggregate visibility into adoption and proficiency — without exposing individual prompts or code.
- ✓Understand team-wide adoption at a glance
- ✓Track team-wide proficiency trends over time
- ✓Identify coaching opportunities across the org
- ✓Measure ROI of your Claude Code investment
Team Overview
Score distribution
Developers see their own scores. Team view shows aggregate data only.
Leaderboard
See how you compare.
Example leaderboard — actual rankings update as developers use the platform.
Your profile is private by default. Appear on the leaderboard only when you choose to.
Trust & Security
Built for developers who care about privacy
You own your data. Export or delete it anytime. Team admins see only aggregate scores — never individual prompts.
No source code collected
The plugin captures coding habits and prompt patterns — never your source code, file contents, or Claude's full responses.
Encrypted in transit and at rest
TLS 1.3 in transit, AES-256 at rest. You can request data export or deletion at any time.
Zero performance impact
Plugin hooks run in the background after operations complete. If our server is unreachable, nothing changes — your Claude Code experience stays fast.
Inspectable plugin
The install script and plugin hooks are plain shell scripts. Read every line before you run anything.
Pricing
Start free. Grow from there.
Individual developers get the full dashboard for free. Teams pay for aggregate insights.
Free
Full dashboard and coaching for individual developers
- ✓10-category proficiency dashboard
- ✓Personalized coaching recommendations
- ✓Overall rating (0–2000)
- ✓Public profile + leaderboard
Team
Aggregate visibility across your engineering org
- ✓Everything in Free for each developer
- ✓Team proficiency dashboard
- ✓Adoption tracking + score trends
- ✓Aggregate reporting
- ✓Org key for easy rollout
Enterprise
Custom onboarding, SSO, and dedicated support
- ✓Everything in Team
- ✓SSO / SAML
- ✓Custom onboarding programs
- ✓Dedicated support + SLA
- ✓Volume licensing
14-day free trial on Team plan · No credit card required · Cancel anytime
15+ seats? Contact us for volume pricing
FAQ
Frequently asked questions
What data does the plugin collect?+
The plugin captures coding habits and prompt patterns — things like whether you use plan mode, how you structure prompts, which tools you invoke, and how you manage context.
It never collects your source code, file contents, or Claude's full responses.
Specifically, the plugin sends: prompt length and structure (not the text), tool names invoked, session duration, plan mode usage, CLAUDE.md presence and structure, commit patterns, and hook configurations. No code, no file paths with content, no Claude output.
Is this a monitoring tool or a coaching tool?+
Coaching. Every developer sees their own proficiency score and personalized recommendations. The goal is to help developers improve.
Team dashboards show aggregate data only — score distributions, adoption rates, and trend lines. Managers never see individual prompts, session logs, or code.
Is it really free?+
Yes. The full 10-category coaching dashboard, proficiency rating, public profile, and leaderboard are free forever. No credit card required.
Paid plans add team dashboards and org-level aggregate insights. Individual developers never need to pay.
I'm an engineering manager. What do I see?+
The team dashboard shows aggregate metrics: average proficiency rating, score distribution across tiers, adoption rate (who has installed the plugin), and week-over-week trends.
You do not see individual prompts, session logs, or per-developer coaching recommendations. Each developer's detailed coaching stays private to them.
You can track whether your Claude Code investment is producing results at the team level without being invasive to individual developers.
What about data retention and deletion?+
Your data is retained while your account is active. You can request a full data export or account deletion at any time by contacting us.
If you unsubscribe from a team plan, your individual data and rating remain accessible on the free tier. Uninstalling the plugin stops all data collection immediately.
I'm new to Claude Code. Is this for me?+
Absolutely. Developers who are just getting started often improve the fastest — there's so much to discover.
prompt1 explains concepts like CLAUDE.md files, plan mode, and efficient prompting as part of your coaching recommendations. You don't need to know these things already.
What is CLAUDE.md?+
CLAUDE.md is a file you place in your project root that tells Claude Code about your codebase — coding standards, architecture patterns, and project-specific instructions.
Think of it as a briefing document for your AI assistant. Developers with a well-written CLAUDE.md get significantly better results from Claude Code. prompt1 helps you write a great one.
Will it slow down Claude Code?+
No. The plugin hooks run in the background after Claude Code operations complete. If our server is unreachable, the hooks silently fail — your Claude Code experience is never affected.
What platforms are supported?+
The plugin works on macOS and Linux. Windows is supported via WSL.
It requires Claude Code CLI. The VS Code extension is not yet supported — we're working on it.
How is the rating calculated?+
Each of the 10 categories is scored 0–100 based on your coding habits. Your overall proficiency rating (0–2000) combines category scores (30%) with historical coding patterns (70%) — things like session consistency, tool diversity, test frequency, and planning habits.
The rating updates as you use Claude Code and build better habits.
What are the rating tiers?+
Five tiers: Bronze (0–499), Silver (500–999), Gold (1000–1499), Platinum (1500–1799), and Diamond (1800–2000).
Most developers start in Bronze or Silver. Applying the coaching recommendations consistently moves you up over time.
How does team onboarding work?+
Team admins get an org key. Each developer runs the install command with the --org-key flag to link their account to the team.
Developers keep full control of their individual profile and coaching data. The team dashboard shows only aggregate metrics.
Find out how good you really are
with Claude Code.
Install the plugin. Get your score. Start improving today.