2 items
Ask about any webpage or PDF and get answers anchored to the source — highlights, notes, citation chasing. Free to try. Onhand is a page-grounded AI tutor for learning and research. Open the side panel on any webpage or PDF, ask a question, and Onhand answers from the material in front of you — and shows you where, highlighting the exact passages it used right on the page. Free to try: no API key, no account. The built-in Onhand Free tier works the moment you install, with a daily usage cap. For more, sign in with OpenAI Codex or bring your own API key (OpenAI, Anthropic, Google Gemini, or any OpenRouter model). WHAT IT DOES — Answers anchored to the source. Every explanation points at the passage that supports it, with highlights and margin notes instead of a detached chat transcript. — Reads PDFs properly. A built-in viewer handles research papers and large documents, with highlights and notes that work the same as on webpages. — Chases citations. Ask "what does [14] actually say?" and Onhand finds the reference, opens the cited paper, and anchors the answer inside it. — Compares sources. Ask how the paper you're reading relates to another open tab, and each claim is highlighted in the source it came from. — Helps you actually learn. Learning mode adds quick comprehension checks as you read and tracks the concepts you've covered, instead of just handing over answers. — Voice mode. Ask aloud while keeping your eyes on the page (uses an OpenAI API key). Study sessions are saved locally in your browser, so you can close a paper and pick the conversation back up later. Privacy note: Onhand starts with the active page. To answer your request it may rank titles and URLs of eligible open tabs and read clearly related open tabs — in any mode, without switching your view; unrelated tabs are not read merely because they are open. In API-key and Codex sign-in modes, relevant request and page content goes to your selected provider. In Onhand Free mode, request content goes through Onhand's hosted Cloudflare Worker to OpenRouter; anonymous diagnostics are required for reliability, quota, cost, and abuse monitoring. Diagnostics and explicit error reports do not include prompts, page content, URLs, screenshots, transcripts, saved sessions, or keys. Onhand is designed for students, researchers, and deep readers who want explanations grounded in the source they're reading — not a chatbot in another tab. Open source: github.com/Phineas1500/Onhand
Aug 2, 2026
rating_count is the Chrome Web Store ratings count, not a written-review count.
Media assets
Screenshots and videos on the listing.
Has promo video
Whether the listing includes at least one video.
Languages
Declared language locales.
Developer website
Listing exposes a developer website URL.
Contact email
Listing exposes a contact email.
Keyword in name
Case-insensitive substring match in the name.
Keyword in description
Case-insensitive substring match in the description.
Keyword occurrences in description
Count of case-insensitive occurrences in the description.
Category user-count percentile
Share of same-category extensions with fewer users (null if unknown).
These are transparent listing completeness / keyword signals, not a prediction of Chrome Web Store search ranking.