Browser automation where an LLM plans and Jev (Typesafe System One) decides. Library, CLI and MCP server.
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.
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.
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 likelydone) 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
From the project README.
Add the radar badge to your README — it shows your project was picked up by MCP Radar and links to this page:
[](https://mcp.liqiwa.com/s/Ying-Kai-Liao--jev-browser.html)
Browser action recorder for automation: record a task once, replay it, or hand the folder to an agent
pairbrowse/pairbrowseMultiplayer browser sessions for developers and AI coding agents
stanislawherjan1/gdocs-comments-mcpMCP server for adding inline, range-anchored comments to Google Docs — the one comment operation the Google Docs & Drive APIs can't do. For AI agents (Claude, Cursor, Copilot) that review documents.
sosoj92/jarvis-assistant-vocalAssistant vocal local en francais : Claude ou Ollama (offline), domotique Hue, OBS, agenda, navigateur, appels Twilio, serveur MCP. Python.
Alpha-Park/genpark-multi-step-web-form-fill-orchestrator-skillMulti-step web form fill orchestrator with validation (Playwright / Puppeteer)
Alpha-Park/genpark-self-healing-dom-semantic-selector-skillSelf-healing DOM semantic selector resilient to class mutations (Tabbit / Stagehand)
The top new MCP servers of the week, every Monday. No spam, unsubscribe anytime.