Can WasItAIGenerated be vibe coded?
Detection API and tools for AI-generated text, images, audio and video, plus a thesis review
The interface is a text box; the product is a fine-tuned classifier. Detection quality does not come from a clever prompt. It comes from training a model on a labelled corpus and tuning it until the error rate is low enough to act on, which is months of work against a target that moves every time a new generator ships. Asking a general model "is this AI-written?" produces confident, unreliable answers, and in an academic-integrity setting that means falsely accusing real students. A one-shot build gets you the UI and none of the responsibility.
Jump to the build brief ↓Legacy-calibrated assessment
Checked Aug 2026
What you pay today, before any DIY hosting
high editorial confidence
Tracked separately from the pricing check
Buildability by layer
Screens, forms, and focused interactions
The repeatable job the product performs
Availability and legality of required data
Uptime, queues, support, and maintenance
Security, compliance, and user confidence
The achievable core
- Build a private workspace that pastes in text, asks one general model for an AI-likelihood judgement with reasoning, stores results, and exports them.
- Wrap a model API in a focused drafting, revision, and export workflow.
- A responsive interface with real empty, loading, success, and error states.
The parts a prompt cannot buy
- a trained classifier and the labelled corpus behind it
- a measured false-positive rate you can quote to an institution
- detection for images, audio and video, not just text
- per-sentence highlighting instead of one document-level guess
- Model quality and inference operations are part of the product.
- The useful dataset is owned, accumulated, or expensive to reproduce.
Build, switch, or keep paying
Narrower, with trade-offs
Build a private workspace that pastes in text, asks one general model for an AI-likelihood judgement with reasoning, stores results, and exports them.
Use the build brief ↓No checked option yet
Compare the prior art below or build only the workflow you need.
Recommended
Institutions do not buy a verdict, they buy a defensible one. A university that flags a student needs a documented error rate, an audit trail and a vendor who will stand behind the number. That is a measurement problem, not an interface problem, and it is why every serious buyer in this category asks about false positives before features.
Visit WasItAIGenerated ↗Why people still pay
Institutions do not buy a verdict, they buy a defensible one. A university that flags a student needs a documented error rate, an audit trail and a vendor who will stand behind the number. That is a measurement problem, not an interface problem, and it is why every serious buyer in this category asks about false positives before features.
Model quality and inference operations are part of the product.
The useful dataset is owned, accumulated, or expensive to reproduce.
Compliance, licensing, and legal accountability are core features.
The brief
Context, requirements, acceptance criteria, non-goals, and the full production standard — as Markdown, ready for any coding agent.
Build brief — a focused alternative to WasItAIGenerated
Context
WasItAIGenerated — Detection API and tools for AI-generated text, images, audio and video, plus a thesis review. It currently costs $9.99/mo.
The interface is a text box; the product is a fine-tuned classifier. Detection quality does not come from a clever prompt. It comes from training a model on a labelled corpus and tuning it until the error rate is low enough to act on, which is months of work against a target that moves every time a new generator ships. Asking a general model "is this AI-written?" produces confident, unreliable answers, and in an academic-integrity setting that means falsely accusing real students. A one-shot build gets you the UI and none of the responsibility.
This brief describes a focused, single-operator replacement for the part of WasItAIGenerated that is genuinely reproducible. It is deliberately narrower than the product it replaces, and it says so in writing. Build the useful core; do not pretend to have rebuilt the rest.
What you are building
Build a private workspace that pastes in text, asks one general model for an AI-likelihood judgement with reasoning, stores results, and exports them.
Wrap a model API in a focused drafting, revision, and export workflow.
A responsive interface with real empty, loading, success, and error states.
Requirements
Functional
Node.js 22.
SQLite database.
A README stating plainly that the output is a guess, not a detector, and must never be used to accuse anyone.
Data and integrations
OpenAI or Anthropic API key in .env.
Each of these needs a real account, credential, or quota. Set them up before writing feature code.
Non-functional
Accessibility: semantic markup, labelled controls, visible focus, and reduced-motion support.
Security: server-side secrets, validated input, and no credentials in the client bundle.
Reliability: retries with backoff on external calls, and a clear failure state when a provider is down.
Portability: the operator can export their data and leave without losing it.
Implementation brief
Build the closest honest consolation tool inspired by WasItAIGenerated. Do not claim it detects AI writing.
Use exactly this stack: Next.js 15 + TypeScript + SQLite.
Primary job: paste text, send it to one general language model, and record that model's opinion on how machine-like the writing reads, with its reasoning.
Start from an empty folder and create the complete working project.
Single-user and private by default; store everything locally.
Put every secret in .env and provide .env.example.
Show the result as an opinion with reasoning, never as a score, a percentage or a verdict - the whole point of this build is that it cannot measure what the paid product measures.
Put a permanent, non-dismissable notice in the UI and the README: this is one model's guess, it has no measured error rate, and it must never be used to accuse anyone of anything.
Include clear empty, loading, validation, success and failure states.
Add export so the user is not trapped in the app.
Accessible keyboard navigation, labels, focus states and sensible contrast.
Deliberately exclude these paid-product advantages: a trained classifier, a labelled corpus, a published false-positive rate, image/audio/video detection, per-sentence attribution.
Do not fake accuracy claims, benchmarks or compliance statements.
Write unit tests for the data model and one end-to-end smoke test of the core loop.
Create a README with setup, architecture, data location and an explicit limitations section.
Run the tests and build before finishing, then fix what fails.
Delivery standard
Inspect the repository first, then write a short implementation plan before writing code.
Deliver the smallest complete end-to-end workflow first; every primary control must work against persisted data.
Use real validation and storage; never substitute fake dashboards, decorative controls, hard-coded success states, or mock integrations.
Include responsive layouts plus genuine empty, loading, success, validation, and failure states.
Keep secrets server-side in environment variables, provide .env.example, and never commit credentials or user data.
Add structured logs around every external call and return actionable errors without leaking sensitive details.
Write unit tests for the core logic and one automated test of the main user journey.
Finish with a README covering setup, architecture, data location, backups, tests, deployment, and known limitations.
Acceptance criteria
A clean install starts the app using only the README and .env.example.
The primary journey works from first visit through saved result, reload, edit, export, and deletion where applicable.
Invalid input, missing configuration, provider failure, and an empty database each have a usable state.
The interface works at 390px and 1440px, is keyboard navigable, and shows visible focus on every control.
Tests, type checking, linting, and a production build all pass with no ignored failures.
No part of the interface implies a live integration, security guarantee, or scale capability that was not actually built and verified.
Non-goals
Do not build these, and do not claim to have replaced them:
A trained classifier and the labelled corpus behind it.
A measured false-positive rate you can quote to an institution.
Detection for images, audio and video, not just text.
Per-sentence highlighting instead of one document-level guess.
Retraining as generators change.
What you still own after launch
Secure credentials, rotate secrets, and handle provider rate limits.
Run migrations, backups, restores, and dependency updates.
Test the critical journey after every model, API, or hosting change.
Monitor failures and fix the edge cases a first prompt will miss.
Risk
High consequence. Use this as a prototype or personal aid. Keep a qualified human and an established provider in the loop for consequential decisions.
Editorial confidence in this assessment: high. No reviewed project implementation is linked yet.
Prior art
Working open-source software you can read, fork, or borrow from before starting:
Binoculars — Zero-shot LLM text detection using perplexity ratios between two models
DetectGPT — Curvature-based zero-shot detection of machine-generated text
Generated by Can It Be Vibe Coded? · Full report: https://www.canitbevibecoded.com/wasitaigenerated
You still own the product
- Secure credentials, rotate secrets, and handle provider rate limits.
- Run migrations, backups, restores, and dependency updates.
- Test the critical journey after every model, API, or hosting change.
- Monitor failures and fix the edge cases a first prompt will miss.
Projects built from this idea
No reviewed implementation has been linked for WasItAIGenerated yet. A submission is evidence for review, not automatic proof that the whole product was replaced.
Built a version of WasItAIGenerated?Submit the project as evidence for this report.
Open-source prior art
Before you start
Can WasItAIGenerated be vibe coded?
Not faithfully. The interface is a text box; the product is a fine-tuned classifier. Detection quality does not come from a clever prompt. It comes from training a model on a labelled corpus and tuning it until the error rate is low enough to act on, which is months of work against a target that moves every time a new generator ships. Asking a general model "is this AI-written?" produces confident, unreliable answers, and in an academic-integrity setting that means falsely accusing real students. A one-shot build gets you the UI and none of the responsibility.
What can an AI coding agent reproduce from WasItAIGenerated?
Build a private workspace that pastes in text, asks one general model for an AI-likelihood judgement with reasoning, stores results, and exports them. Wrap a model API in a focused drafting, revision, and export workflow. A responsive interface with real empty, loading, success, and error states.
What will a DIY WasItAIGenerated replacement still be missing?
a trained classifier and the labelled corpus behind it; a measured false-positive rate you can quote to an institution; detection for images, audio and video, not just text; per-sentence highlighting instead of one document-level guess; Model quality and inference operations are part of the product.; The useful dataset is owned, accumulated, or expensive to reproduce.
What do I still own after building a WasItAIGenerated alternative?
Secure credentials, rotate secrets, and handle provider rate limits. Run migrations, backups, restores, and dependency updates. Test the critical journey after every model, API, or hosting change. Monitor failures and fix the edge cases a first prompt will miss.