covateOpen-source MCP · free learning ledger
Turn AI-assisted coding into real understanding.
When AI writes most of your code, it’s easy to ship things you don’t actually understand. Covate quizzes you on your own changes so you learn while you build — then, if you sign in with GitHub, syncs every session into a learning ledger you can read back. It is free; there is nothing to buy.
Free · no card, no waitlist
MIT-licensed MCP · free learning ledger · no paid tier
“It works… but could you explain it?”
The faster AI writes your code, the quieter your own skills erode. You accept a diff, it passes the tests, you move on — and six months later you couldn’t rebuild it from scratch. Covate is the antidote: a lightweight loop that checks your understanding at the exact moment you still have the context, so using AI makes you faster and better, not just faster.
How it works
Code with your AI assistant
Keep using Claude, Cursor, Copilot — whatever you already use. Covate rides along as an MCP server, watching what actually changes in your codebase.
Get quizzed on what just changed
Ask Covate to quiz you (or set a cadence) and it turns the recent diff into a few sharp questions about the concepts, APIs and trade-offs — blocking further generation until you answer. You learn while the context is fresh.
Watch your understanding compound
Every session is saved locally. Sign in with GitHub — free — and the sync client pushes them into your learning ledger, where you can read back each session, see your running accuracy, and find the topics you answer worst.
What you get, all of it free
Two halves, both free: an MIT-licensed MCP that runs on your machine, and a hosted learning ledger you open by signing in with GitHub. There is no paid tier and no checkout anywhere on this site.
Free · open-source
The Covate MCP
- learning_session: quizzes from your real code changes
- debug_search: project-level debug memory (RAG)
- Local session history + local quiz WebUI
- Runs fully offline, no account, MIT-licensed
Free · live now
The learning ledger
- Sign in with GitHub — free, no card, no waitlist
- Your synced sessions, newest first, with per-session scores
- Totals: sessions, questions, correct answers, accuracy
- The topics you answer worst, ranked
- Your sync token: reveal it or rotate it whenever you want
No charts over time, no study planner, no teams yet — and nothing to buy.
Sign in with GitHub ->How can you check that you actually understood the code?
Five things people do after an AI writes a change, compared on what each one actually proves rather than on how thorough it feels.
| Way to check | What it proves | Stops the AI mid-flow? | Needs another person? | Leaves a record? | Cost |
|---|---|---|---|---|---|
| Read the diff yourself | Nothing you can check later — recognising code reads as understanding it | No | No | No | Free |
| Ask the AI to explain it | That the AI can produce an explanation, not that you followed it | No | No | In the chat log only | Free |
| Human code review on the pull request | A second person's reading — 45% of developers say reviewing and debugging AI code takes longer | No — it happens after the code exists | Yes | Yes, review comments | A reviewer's time |
| Write tests for the change | The behaviour you thought of; AI-written tests can pass tautologically | No | No | Yes, the test suite | Free |
| Answer questions on your own diff (Covate) | Whether you can answer 3 questions about the change you just accepted | Yes — the tool returns HALT_GENERATION until you answer | No | Yes, the learning ledger | Free, MIT |
The 45% figure is from the 2025 Stack Overflow Developer Survey; the Covate row is read from this repository's source, not estimated.
What does the research say about AI-written code?
Four primary sources, each linked below, and every figure quoted from the source rather than summarised second-hand. They disagree with each other in useful ways.
Does AI actually make developers faster?
Not always. In a 2025 randomized controlled trial, 16 experienced open-source developers predicted AI would cut task time by 24% and afterwards believed it had saved 20%. Measured, the same 246 tasks took 19% longer with AI than without. METR, arXiv:2507.09089
Why do other studies show a large speed-up?
Because they measure different work. A 2023 controlled experiment found developers with GitHub Copilot finished a self-contained HTTP-server task 55.8% faster. That is greenfield code; the 2025 trial used mature repositories the developers already knew well. arXiv:2302.06590
How much do developers trust what AI writes?
Less every year. In the 2025 Stack Overflow Developer Survey, 84% of respondents use or plan to use AI tools, up from 76% in 2024 — yet 46% distrust the accuracy of the output against 33% who trust it, and only 3% highly trust it. Favourable sentiment fell from over 70% in 2023 and 2024 to 60%. Stack Overflow, 2025
Where does AI-generated code actually go wrong?
The same survey ranks the frustrations: 66% cite “AI solutions that are almost right, but not quite”, 45% say debugging AI-generated code takes longer, 20% report losing confidence in their own problem-solving, and 16.3% say it is hard to understand how or why the code works.
Is AI-assisted code less secure?
In a Stanford user study, participants with an AI assistant wrote significantly less secure code than those without — and were more likely to believe their code was secure. Read every generated diff that touches input, auth or secrets against the OWASP Top 10 and the CWE Top 25. arXiv:2211.03622
What exactly does Covate install and do?
Everything below is read from the open-source repository, so you can check each number against the code before you install anything.
What does the MCP server actually add?
4 tools: learning_session (quizzes you on recent changes), debug_search and debug_record (a per-project debugging memory), and term_get (concept lookups across 10 domains). It needs Python 3.11+, binds to 127.0.0.1, and is MIT-licensed.
How long does one learning session take?
A session defaults to 3 questions with 4 options each, generated from 7 change-type templates and aimed at 5 focus areas — logic, security, performance, architecture and syntax. The default cap is 10 minutes, configurable from 1 minute to 2 hours.
Where do the sessions end up?
On your machine first: the server keeps the 100 most recent sessions per project, and the quiz UI ships in 3 languages. Signing in with GitHub and running the sync client pushes them into the learning ledger, which is free — there is no paid tier anywhere on this site.
Does it work with my assistant?
If your assistant speaks the Model Context Protocol, yes — Claude, Cursor, Copilot and the rest all load MCP servers the same way, described in the 2025-06-18 specification.
Sources
Every figure on this page comes from one of these, or from the Covate repository itself. Each link was checked on 27 August 2026.
- Model Context Protocol — specification (2025-06-18)The protocol Covate implements. Defines how a tool is exposed to an AI assistant and how a call returns.
- METR — Measuring the Impact of Early-2025 AI on Experienced Open-Source Developer ProductivityRandomized controlled trial, 16 developers, 246 tasks. Source of the 19% slowdown and the 20% / 24% self-estimates.
- Stack Overflow Developer Survey 2025 — AI sectionSource of the 84% adoption, 46% distrust, 66% “almost right” and 45% debugging figures.
- DORA — State of AI-assisted Software Development 2025Google Cloud’s annual study; finds AI amplifies an organisation’s existing strengths and weaknesses rather than replacing them.
- OWASP Top 10The 10 web-application risk categories to read a generated diff against before merging it.
- CWE Top 25 Most Dangerous Software WeaknessesMITRE’s ranked list of the 25 weakness classes that cause the most real damage.
- Git — git-diff documentationThe diff Covate reads is an ordinary git diff; the official flags for narrowing it are here.
Questions
What is Covate?
Covate is an open-source learning sidecar for AI coding assistants. Its MCP tool, learning_session, watches your recent code changes and turns them into short, interactive quizzes that pause the AI until you've understood what just got built — so you learn while you ship instead of blindly accepting generated code. Sign in at covate.org with GitHub and the sync client pushes those sessions into a learning ledger you can review in the browser.
How does it work?
You code with your AI assistant as usual. When you ask Covate to quiz you (or on your own schedule), it reads the recent changes and generates a few targeted questions about the concepts, APIs, and trade-offs involved, blocking further generation until you answer. Each session — the questions, your answers, the topics — is saved locally. If you sign in with GitHub and run the sync client with your sync token, those records land in your learning ledger on covate.org, where you can review every session, your running accuracy, and the topics you keep getting wrong.
Is it really free?
Yes, all of it. The Covate MCP server is open-source (MIT), runs fully locally and needs no account. The learning ledger on covate.org is free too: sign in with GitHub, sync, review. There is no paid tier, no subscription and no checkout — nothing on this site can be bought.
What does signing in add?
One thing, honestly described: your local learning sessions become reviewable in a browser. The ledger shows every synced session, how many questions you got right, your overall accuracy, and a ranked list of the topics you answer worst — plus your sync token, which you can reveal or rotate at any time. It is not a study planner: it does not schedule reviews or re-quiz you on a spaced-repetition timetable today.
Why do I need this if the AI writes the code for me?
That's exactly the problem it solves. When an AI writes most of your code, it's easy to ship things you don't actually understand — and your own skills quietly erode. Covate keeps you in the loop: it makes sure you understand each change while you still remember the context, and the ledger keeps a record so you can see which concepts you keep missing. It's the difference between using AI to move faster and using AI to also get better.
What does it cost, and what's coming next?
Nothing. The MCP is available now from the repo, and the learning ledger is live and free — sign in with GitHub at covate.org/dashboard. Progress-over-time charts, spaced-repetition study plans and team accounts are ideas, not shipped features; when they exist we'll say so here, and if any of them ever becomes paid you'll see a price before you're asked for anything.
Learn from every change.
Install the open-source MCP, sign in with GitHub, and every session you run shows up in your ledger. Free, today.
Free · no card, no waitlist