Artifacts

Upload a PDF and Claude writes a flashcard deck from it, which you then practice through games, quizzes and an XP system.

Assessment that scores how mature an organization's AI adoption is, using Claude for the analysis.

Italian-language tool: load an accounting document, ask questions or generate reports on IFRS statements with your own Anthropic key.

Turns a master resume and a job description into fit scores and apply-ready application drafts.

German-language fact checker: paste a claim or news excerpt and an AI model assesses its truthfulness using your Anthropic key.

Expands a short feature idea into a structured spec for developers, including user stories and requirements.

Photographs a business card, extracts contact fields with AI, adds background notes and follow-up email drafts, and exports a vCard QR code.

Turns a photo into a rhythm-synced glitch or electro-cutout flicker video, with AI subject segmentation for social posts.

A local-first block notes app with nested pages, templates and an optional AI assistant using a key you supply.

Extracts text from PDFs and images with Tesseract.js; the page says files stay in your browser.

Runs prompts against the language model built into Chrome, downloading it on first use.

Drop in a picture or use the webcam and a CLIP model run by Transformers.js says whether a bird is in it.

Blurs faces and strips metadata from a photo locally, then saves a clean copy.

Counts tokens for text and image attachments across several Claude models using Anthropic token-counting API and your key.

Chat client that connects from the browser to any OpenAI Responses-compatible endpoint, such as OpenRouter.

Asks Gemini to locate objects in an image and draws the returned bounding boxes on top.

Bare-bones chat page that talks to a selectable Gemini model using your own Google API key.

Simple single- or multi-speaker speech generator for the Gemini Flash TTS model with directorial tags.

Requests segmentation masks from Gemini for objects in an image and overlays them for inspection.

Talk to a Gemini Live model by voice in the browser, with transcripts and optional pasted document context.

Send a system prompt plus content to a chosen Gemini model and read the response, with your own API key.

Compose single-voice or conversation scripts, set delivery style and voices, and render speech with Gemini TTS.

Front end for the Jina Reader API that turns a URL into Markdown or text, with an optional Claude summary step.

Try one JavaScript library against OpenAI, Anthropic or Gemini models by picking a provider and pasting a key.