lgtm? – Jev-powered checks that make agents test 是一个「skill」类 AI agent,解决:在构建 AI 代理时自动验证其生成代码的正确性与安全性。定价:$0.042/百万输入tokens。截至 2026-10-27,KanonAgent 记录到 2 票。本站首次收录于 2026-09-20。
一个由 Jev 驱动的检查系统，帮助 AI 代理自动验证代码质量与行为，解决开发者在构建 AI 代理时缺乏可靠测试手段的痛点，适合 AI 工程师与自动化开发团队使用
2 票
Kanon 于 2026 年 9 月 20 日 收录
状态依据:npm_downloads_7d gained 6 in the last 3d after a flat 3d baseline (no prior movement to compare a ratio against)
📈 时间线什么时候变了什么
- 加速 npm_downloads_7d gained 6 in the last 3d after a flat 3d baseline (no prior movement to compare a ratio against)
- 加速 npm_downloads_7d gained 6 in the last 3d vs 2 in the 3d before (ratio 3.0x, threshold 1.5x)
- 加速 npm_downloads_7d gained 6 in the last 3d vs 2 in the 3d before (ratio 3.0x, threshold 1.5x)
- 加速 npm_downloads_7d gained 6 in the last 3d vs 2 in the 3d before (ratio 3.0x, threshold 1.5x)
- 降温 npm_downloads_7d momentum fell to -3 per 3d from 11 (<= 40% of the prior window)
- 降温 npm_downloads_7d momentum fell to -3 per 3d from 11 (<= 40% of the prior window)
- 降温 npm_downloads_7d momentum fell to -12 per 3d from 11 (<= 40% of the prior window)
- 降温 npm_downloads_7d momentum fell to -12 per 3d from 11 (<= 40% of the prior window)
- 降温 npm_downloads_7d momentum fell to -1217 per 3d from 217 (<= 40% of the prior window)
- 降温 npm_downloads_7d momentum fell to -568 per 3d from 205 (<= 40% of the prior window)
- 异常 daily change -580.0 is -3.1 standard deviations from this entity's own prior mean (143.7±232.3, baseline excludes this point) — unusual, not necessarily adoption
- 降温 npm_downloads_7d momentum fell to -391 per 3d from 28 (<= 40% of the prior window)
- 异常 daily change -580.0 is -3.1 standard deviations from this entity's own prior mean (143.7±232.3, baseline excludes this point) — unusual, not necessarily adoption
- 降温 npm_downloads_7d momentum fell to 217 per 3d from 645 (<= 40% of the prior window)
- 加速 npm_downloads_7d gained 217 in the last 3d after a flat 3d baseline (no prior movement to compare a ratio against)
- 异常 daily change -580.0 is -3.1 standard deviations from this entity's own prior mean (143.7±232.3, baseline excludes this point) — unusual, not necessarily adoption
- 降温 npm_downloads_7d momentum fell to 205 per 3d from 645 (<= 40% of the prior window)
- 首次观测 Kanon 开始为它保留永久历史
🔎 已知 / 未知
23 个字段里 13 个未知
| 可机器调用 | — | 未知 | — |
| 开源 | 1 | 推断 | 1 条引文 |
| 可自托管 | — | 未知 | — |
| 需自带 key | 1 | 推断 | 1 条引文 |
| 自主度 | 2 | 推断 | 1 条引文 |
| 计费模式 | — | 未知 | — |
| 集成面 | github | 推断 | 2 条引文 |
「未知」= 我们没核实过,**不等于「否」**。硬性筛选永远不会把未知当成否。
🤖 Agent 拆解 · skill
解决什么场景在构建 AI 代理时自动验证其生成代码的正确性与安全性
自主度L2 · 有工具调用(引文: "It runs on [Jev by TypeSafe](https://typesafe.ai), with your…")
前置条件开源 · 需自带 API key · 仅厂商云
集成面github
定价$0.042/百万输入tokens
为什么值得关注
通过集成 Jev 实现对 AI 代理行为的动态验证，是 AI agent 可靠性建设的关键补位
证据引文判定所依据的原文 · 逐字引用
“public GitHub repository (README fetched)”— structural
“It runs on [Jev by TypeSafe], with your own `TYPESAFE_API_KEY`.”— readme
“It asks for your TypeSafe API key (get one at <https://typesafe.ai>)”— readme
“It runs on [Jev by TypeSafe](https://typesafe.ai), with your own `TYPESAFE_API_KEY`.”— readme
“github.com/stardeckai/lgtm”— readme
“github/actions/workflow”— readme
信号来源: Show HN
访问官网 →
📛 官方徽章挂到官网 / README
你是 lgtm? 的作者?上面选样式,嵌入代码实时更新;要贴合站点配色可在 URL 上加 bg= / fg= / accent=(十六进制色)。徽章链回本页。
手机端点「分享」直达微信/朋友圈/小红书;桌面端用「复制文案」后到 App 内粘贴发布
常见问题
lgtm? 是什么?
一个由 Jev 驱动的检查系统，帮助 AI 代理自动验证代码质量与行为，解决开发者在构建 AI 代理时缺乏可靠测试手段的痛点，适合 AI 工程师与自动化开发团队使用
lgtm? 解决什么问题?
在构建 AI 代理时自动验证其生成代码的正确性与安全性
lgtm? 为什么值得关注?
通过集成 Jev 实现对 AI 代理行为的动态验证，是 AI agent 可靠性建设的关键补位
lgtm? 多少钱?
$0.042/百万输入tokens
lgtm? 是开源的吗?
是。lgtm? 为开源项目。
lgtm? 能接哪些工具?
github
lgtm? 有多少人在用?
KanonAgent 记录到:2 票(本站首次收录于 2026-09-20)。
lgtm? 有什么替代品?
KanonAgent 库内的同类 agent:Stealth Venture、taste-skill、archify、open-design、deer-flow、ruflo。