Obscura – V8-powered headless browser for scraping and AI agents
30MB memory and instant startup vs Chrome's 200MB and 2-second load.
It runs a headless Chromium to bypass Cloudflare, extracts consistent product fields, caches results, and emits clean Markdown so an LLM agent can do multi-step product comparisons — and it even ships as a Claude Code skill for autonomous queries. Smart, practical engineering, but the whole thing rests on brittle scraping (Cloudflare/site changes) and the README could use clearer notes on rate limits, legal/ToS tradeoffs, and agent integration examples.
AI/agent developers, biohackers and supplement researchers, and developers needing structured iHerb product data
The CLI outputs clean Markdown, caches results, and handles Cloudflare bot protection via a headless browser. It ships as a Claude Code skill, so you can ask things like "compare the top 3 magnesium glycinate supplements by price and bioavailability" and the agent will do multi-step research on its own.
Built in Rust. No API key or account needed.
30MB memory and instant startup vs Chrome's 200MB and 2-second load.
Headless spreadsheet SDK for agents when SheetJS and HyperFormula already exist.
Headless PTY daemon lets AI agents control ncurses apps like k9s without a GUI.
Real browsers beat bot detection where Puppeteer and Playwright fail—you rent idle machines.
Smartly checks for native markdown files before falling back to HTML scraping.
No Chromium, no pixels—just semantic trees agents actually need in under 200ms.