jev-browser

Browser automation where an LLM plans and Jev decides.

Unofficial project, not affiliated with TypeSafe. It calls the TypeSafe System One API with your own API key.

The calling LLM (Claude, via MCP) says what outcome it wants, one step at a time, and hands over any text to type. For each round of a step, code describes the page. Then one ~300 ms Typesafe System One request asks Jev several questions at once: which element, which action, which value, and is the step done / blocked / showing an error / about to do something irreversible. Playwright performs the action. The LLM never reads page snapshots unless it chooses to take over.

Claude ── browser_do("Log in", {email, password}) ──▶ jev-browser
                                                       │  loop until done / stuck / needs confirmation
                                                       │   1. settle   (network + DOM quiet)
                                                       │   2. describe (elements, labels, state, visible text, diff, counts)
                                                       │   3. Jev      (done? error? irreversible? tool? target? value?)
                                                       │   4. act      (Playwright)
Claude ◀── { status: "done", url, actions[], done_score } ─┘

https://github.com/user-attachments/assets/2e688df9-4985-4854-8ebe-ba97c9d13d68

Jev only answers with probability distributions: yes/no (noul), pick one option (choice) or a rating (score). It never writes text. So everything free-form comes from the caller as candidates, and code turns disagreement or low confidence into a status the LLM can act on.

Results

42 tasks in 16 categories on live sites (see RESULTS.md):

• 40/42 correct in the latest run, 0 false "done" claims (38/41 twice before the latest fixes). Remaining misses: counting ("add until 3", flagged likely_done) and verifying a sort.
• ~300 ms per Jev call, 2–4 calls for most steps; a 5-step checkout takes ~14 s end to end.
• Pause before irreversible actions: across ~200 rounds it flagged only saucedemo's "Finish" (place order) button.
• Less page content for the LLM than a Playwright-MCP-style loop on the same tasks (estimate): median 5× per task, ≈8k vs ≈557k tokens in total. Big pages dominate the total: a Wikipedia article is ~149k snapshot tokens. On tiny pages there is no saving. The page reading moves to Jev.

Works on: forms, native and custom dropdowns, checkboxes and radios (including styled replacements), dynamic loading, modals, JS dialogs, hover, right-click, drag and drop, key presses, file upload, iframes, shadow DOM, new tabs, pages with 2,000+ elements (two-stage selection), and non-English UIs.

Known limits, so write steps around them:

• Ordered sub-goals in one step ("add two todos, complete one, clear completed") → split into one outcome per step.
• Open-ended goals ("scroll to load more") → make them measurable or check yourself.
• Judgements that compare many values (is this table sorted, did exactly one thing change) → verify with browser_check or browser_snapshot.

Quick start (MCP, from npm)

npx playwright install chromium          # once
claude mcp add jev-browser -e TYPESAFE_API_KEY=your-key -- npx -y -p jev-browser jev-browser-mcp

Any MCP client works the same way: command npx, args -y -p jev-browser jev-browser-mcp, env TYPESAFE_API_KEY. Add JEV_BROWSER_HEADED=1 to watch it work.

Setup (from source)

git clone https://github.com/Ying-Kai-Liao/jev-browser && cd jev-browser
npm install
npm run setup                    # downloads Chromium for Playwright
cp .env.example .env             # add TYPESAFE_API_KEY
npm test                         # offline tests (no network, no key)
npm run test:e2e                 # MCP server end to end (network + key)

Use from Claude Code (MCP, from source)

claude mcp add jev-browser -- node /absolute/path/to/jev-browser/bin/jev-browser-mcp.mjs

From a source checkout the server reads TYPESAFE_API_KEY from the repo's .env.

Reading a page: browser_read

browser_snapshot is the take-over-and-act surface: numbered elements, plus the text in the viewport. That is the right trade for acting, and it is already small (99–6,900 tokens on every page measured, because elements are capped at 400 and the text at 2,500 chars). It is a poor way to read, though — a Wikipedia article is ~179k chars of text, so a snapshot shows about 1.4% of it and the rest costs a scroll-and-snapshot loop.

browser_read(question) is the understand-the-content surface. Code extracts the whole document's text and cuts it at headings and sections; Jev rates every block against the question in batches of 40; code returns the blocks that help, in document order, inside a character budget. Jev ranks, it never writes — the text you get back is the page's own.

When it pays. Long articles, documentation and reference pages. On the WW2 article, "When and why did Japan surrender?" returned 11.8k of 178.5k chars — 15× less — in 3.4 s over 5 Jev calls, keeping the atomic-bomb and surrender paragraphs.

When it doesn't. Short pages, apps and forms. Below a 8,000-char floor (and below your own max_chars) it returns the page whole and never calls Jev at all, because paying ~1 s to prune a 2k-token page is a loss. Hacker News, a BBC front page and saucedemo all land under the floor.

It is biased to keep: a block Jev cannot rate — or a whole batch whose request failed — is kept, and whatever is left out is counted and marked in place with [… N blocks dropped …]. Nothing is ever silently cut. Use browser_snapshot if you need the page as the caller normally sees it.

node examples/read-demo.mjs                                   # WW2 article, default question
node examples/read-demo.mjs "https://…" "what do you want?"   # needs a key

browser_do statuses:

browser_snapshot options. A snapshot is the one page the calling LLM reads in full, so every option narrows what gets rendered. All of them are pure formatting: nothing is collected differently and none of them costs a Jev call.

A typed password is never printed: input:password values render as [redacted], in a snapshot and in a diff. Measured on a Wikipedia article (1,840 elements), a default snapshot went from 26,919 to 9,926 characters (~6,700 → ~2,500 estimated tokens); filter brings it to ~800 and diff after a scroll to ~50.

Env: JEV_BROWSER_HEADED=1 shows the browser, JEV_BROWSER_PROFILE=/dir keeps a persistent profile (logins survive restarts), JEV_BROWSER_LOG=1 prints per-round decisions to stderr.

Library

// npm install jev-browser &&