Pairing Codex with Jev for browser automation: 5-10x speedup with a split-labor workflow
alexcovo_eth · x · 2026-09-19
An open-source workflow pairs Jev (browser-control agent) with Codex for web automation, claiming 5-10x speedups by splitting labor: Jev continuously executes clicks, jumps, and scrolling, while Codex only steps in for typing and image-based judgment—cutting model round-trips, latency, and token usage.
Setup steps:
- Get Codex (with browser/CUA), Node.js 22+, TypeSafe API key
- Install the skill via the repo or npm run install:codex -- --env-file .env
- Load jev-use so Jev takes over pointer/navigation actions
- Hand back to Codex for judgment/input, then let Jev resume
- Validate against real product pages or automation tests
Bonus use case: Jev's Instant Compaction scores each tool call KEEP/DROP to compress a 156k-token Claude Code session to 62k, preserving verbatim errors and code instead of lossy summaries.
Related event: Open-Source Jev-cu Gives Codex Faster Computer Use Capabilities(2 posts)→
More from coding & agent
- Why LLM-Written Docs Fail: Models Optimize for Punchiness, Not Readability — mervenoyann · 2026-11-04
- OpenAI DevDay demo shows forking agent task history at any step — paw_lean · 2026-11-04
- Veteran game dev: "AI builds a full game in a day" posts are mostly hype — bendee983 · 2026-11-04
- Dev builds interactive Excel-to-dashboard with ChatGPT, shares the prompt — TermRemarkable665 · 2026-11-04
- Agents cut dev time from weeks to a day; bottleneck moved to specs and review — Suspicious_Orchid770 · 2026-11-04
- Orgo, a YC-backed platform giving AI agents their own cloud desktops, claims 1,000+ teams — Scobleizer · 2026-11-04