No red flags found in any of the 11 categories — no credential harvesting, no data exfiltration, no curl-pipe-shell installer. Scanned against the SlowMist agent-security taxonomy, refreshed every 8 hours. Full audit →
by YUTA-fywoo · Codex Skill · ★ 131
Last updated: · Indexed by AgentSkillsHub · Auto-synced every 8h
🔒 Is jev-gui-delegate safe to install? View the security audit →
Jev GUI Delegate 面向 Codex 的 Windows / Chrome GUI 委派原型。主模型交付任务与验收条件，本地控制器执行确定性步骤，需要语义判断时调用 TypeSafe Jev，并在每一步核对实际结果。 这是个人设计、借助 AI 实现并持续验收的工具项目。项目贡献包括任务契约、工具编排、低置信恢复、权限与状态检查、测试和反馈迭代；Jev 模型、TypeSafe SDK、Codex 和浏览器插件属于上游产品。 为什么做 逐步操作网页会让主模型反复读取界面、选择控件和接收工具返回。这个项目把一段可明确验收的操作交给本地控制器，减少主模型在操作循环中的参与，同时保留状态检查、异常恢复和用户接管。 已完成的设计 两条执行路径：Chrome 复用宿主提供的官方浏览器会话；Windows 使用 UI Automation。没有复制个人浏览器配置或提取 Cookie。 可检查的任务契约：目标、允许动作、范围、输入、步骤和成功条件显式定义。 按需使用模型：确定性匹配由本地代码执行；语义选择调用 Jev。零模型调用按实际记录，不当作 Jev 推理成功。 连续执行与恢复：保存进度，支持缺少输入的补充、局部选择修复、暂停、取消和用户接管；结果不明时不自动重放提交。 结果核验：操作前核对身份和范围，操作后检查实际值或页面状态；使用量与失败原因分开记录。 对照测试 2026-09-23 的两轮合成任务记录，每轮都比较同样的 4 项任务、24 个动作。下表统计主模型 Astra 的输入和输出 token，包含缓存输入；Jev 的额外用量单列在测试说明。 这是固定任务、有限样本的操作循环测试。两轮重复相同四项任务，不代表八种独立场景；结果不是账单节省比例，也不代表所有应用、任意任务或部署成本。早期三任务版本的非缓存 token 结果并未改善，详见测试说明。 阅读源码 | [gui
| Stars | 131 |
| Forks | 6 |
| Language | Python |
| Category | Codex Skill |
| Quality Score | 57.4708172657435/100 |
| Last Updated | 2026-09-27 |
| Created | 2026-09-25 |
| Platforms | browser, codex, python |
| Est. Tokens | ~14k |
These tools work well together with jev-gui-delegate for enhanced workflows:
Looking for a jev-gui-delegate alternative? If you're comparing jev-gui-delegate with other codex skill tools, these 3 projects are the closest alternatives on Agent Skills Hub — ranked by topic overlap, star count, and community traction.
开源了Codex Skills等集合，大部分是自己实际需要搞得，需要的自取。欢迎star
Skills for agentic development on Solana
A Codex/Claude skill that maintains per-repo operating theory documents.
Explore other popular codex skill tools:
jev-gui-delegate is AI-assisted Windows and Chrome GUI delegation for Codex: task contracts, local execution, Jev semantic decisions, recovery and outcome verification.. It is categorized as a Codex Skill with 131 GitHub stars.
jev-gui-delegate is primarily written in Python.
You can find installation instructions and usage details in the jev-gui-delegate GitHub repository at github.com/YUTA-fywoo/jev-gui-delegate. The project has 131 stars and 6 forks, indicating an active community.
The top alternatives to jev-gui-delegate on Agent Skills Hub include skills, solana-dev-skill, theorist. Each offers a different approach to the same problem space — compare them side-by-side by stars, quality score, and community activity.
Grades come from a rule-based scan built on the SlowMist agent-security taxonomy, covering 11 red-flag categories including credential harvesting, data exfiltration, and curl | sh installers. It is a first-layer scan, not a manual audit — we say so rather than overstate it.
The scale of the problem is documented independently: Liu et al. (2026), in a study of 31,132 agent skills, report that 26.1% contain security vulnerabilities. Our own full-catalog census is published as a citable open dataset.
Sources & who's responsible: