
Aihot
Pull today’s curated Chinese AI industry brief from AI HOT without browser tabs or stale model cutoff dates.
Overview
aihot is an agent skill most often used in Idea (also Validate, Operate) that fetches live AI HOT briefings from aihot.virxact.com and summarizes them in Chinese markdown for agents.
Install
npx skills add https://github.com/kkkkhazix/khazix-skills --skill aihotWhat is this skill?
- Curls public REST endpoints on aihot.virxact.com—no API key or MCP server required
- Formats responses as Chinese markdown briefings (daily HOT, featured picks, weekly papers)
- Mandatory browser User-Agent on /api/public/* to avoid nginx 403 on default curl UA
- Install path /aihot-skill/ exempt from UA blacklist for one-line curl | bash setup
- Explicit anti-undertrigger: must invoke for any Chinese or English “AI news today” style query
Adoption & trust: 992 installs on skills.sh; 14.1k GitHub stars; 2/3 security scanners passed (skills.sh audits).
What problem does it solve?
You need today’s AI industry news in Chinese but asking the model alone mixes old training data with guesses.
Who is it for?
Solo builders and indie founders who want agent-native AI news pulls during research standups or before committing to a feature bet.
Skip if: Teams that need paywalled Western wire services, deep forensic source verification inside the repo, or offline-only workflows with no outbound network.
When should I use this skill?
User asks for AI HOT, AI 日报/资讯/热点, today’s AI news, recent OpenAI/Anthropic/Google releases, or any AI-circle happenings—including vague “what happened today” in an LLM context.
What do I get? / Deliverables
You get a structured, source-backed markdown briefing from the public AI HOT API after successful curl calls with the required User-Agent header.
- Chinese markdown briefing of AI HOT entries
- Structured summary with links from public API JSON
Recommended Skills
Journey fit
Spans multiple journey phases - primary shelf plus alternate fits below.
Canonical shelf is Idea because the skill’s primary job is answering “what happened in AI today” for discovery and positioning before you build. Research fits daily briefings on model releases, products, papers, and industry moves sourced from live API data.
Where it fits
Pull the daily HOT list before drafting a positioning doc on a new model API.
Check whether a competitor product launch this week changes your MVP cut line.
Verify which provider shipped an endpoint you plan to wire into your agent this sprint.
Friday scan of featured entries to decide if a dependency upgrade is urgent.
How it compares
Use instead of trusting the model’s memory for “AI news today”—this is a live HTTP integration, not general web search.
Common Questions / FAQ
Who is aihot for?
Indie and solo builders using agent IDEs who read or write Chinese and want curated AI industry updates inside the coding session.
When should I use aihot?
In Idea research for daily HOT and model-release scans; during Validate when scoping bets against recent launches; in Operate when monitoring whether a dependency or competitor shipped something yesterday.
Is aihot safe to install?
It only calls documented public endpoints over HTTPS; review the Security Audits panel on this Prism page and your org policy on external news APIs before enabling network in agents.
SKILL.md
READMESKILL.md - Aihot
# AI HOT Skill 让 Agent 用最自然的中文查询拿到 aihot.virxact.com 上每天的 AI HOT 日报和全部 AI 动态,不需要打开浏览器。SKILL.md 标准格式,跨 Claude Code / Codex CLI / Cursor / Gemini CLI / OpenCode / 任何兼容平台可用。 线上:https://aihot.virxact.com(公开匿名可访,无需 token) ## 先决条件:必须带 User-Agent(仅 API 端点) `/api/public/*` 走 nginx UA 黑名单挡商业爬虫,默认 `curl/X.Y` UA 会被 403 Forbidden。**调 API 时所有 curl 都必须带浏览器 UA**: ```bash UA="Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/124.0.0.0 Safari/537.36" # 之后所有调 API 的 curl 都加 -H "User-Agent: $UA",例如: curl -sH "User-Agent: $UA" "https://aihot.virxact.com/api/public/daily" ``` 后面"工作流"章节的 curl 例子为了简洁默认你已经设了 `$UA`——实际调用必须加 `-H "User-Agent: $UA"`,**不要忘**。漏掉这一步会让你以为接口挂了,实际只是被 403 挡了。 > **范围澄清**:这条 UA 要求**只针对 `/api/public/*` API 端点**。`/aihot-skill/{install.sh,SKILL.md,README.md}` 安装入口 nginx 上**特意豁免** UA 黑名单(设计前提就是给 `curl -fsSL ... | bash` 一行装用),用 default curl UA 直通 200。不要把"先决条件"误推广到所有 aihot.virxact.com 路径。 ## 什么时候用 > **路由优先级(第一原则)**:**默认走精选** `items?mode=selected`——它是 AI HOT 每天精挑细选的"主菜单",覆盖用户关心的事且数据新鲜。 > > - **仅当用户在话里明确说出"日报"** 二字才走 `daily`(编辑成品,按 UTC 整日切片,跟"过去 24 小时 / 今天"等滚动窗口对不上) > - **仅当用户明确说"全部 / 完整 / 所有 / 全量"** 才走 `mode=all`(含未精选的次要条目,量大但杂) > - **"今天 AI 圈"、"过去 24 小时大新闻"、"最近 AI 圈有啥"** 等宽问题 = **默认精选 + 时间窗(since)**,不要默认走日报或全部 > > 这是为了对齐用户的语义优先级:精选是主菜单,日报和全部是用户特意点单的备选,不应抢默认。 | 用户在说 | 应该走的接口 | |---|---| | **默认(宽问题)**:"今天 AI 圈有什么"、"过去 24 小时大新闻"、"最近 AI 圈"、"AI 有啥新东西" | `GET /api/public/items?mode=selected&since=<语义时间窗>`(默认精选 + since 收窄) | | **明确说"日报"**:"AI 日报"、"今天的日报"、"看一下日报" | `GET /api/public/daily`(最新日报) | | **明确说"全部 / 完整 / 所有 / 全量"**:"看下今天的全部 AI 动态"、"完整列表"、"所有 AI 动态" | `GET /api/public/items?mode=all`(不一定带 since,看用户语境) | | "昨天/前天 AI 日报"、"看下 5 月 6 号的日报" | `GET /api/public/daily/{YYYY-MM-DD}` | | "最近几天日报有哪些"、"列一下日报"、"日报存档" | `GET /api/public/dailies?take=N` | | "看下精选条目"、"AI HOT 精选" | `GET /api/public/items?mode=selected` | | "最近的模型发布"、"AI 产品发布"、"AI 行业动态"、"AI 论文" | `GET /api/public/items?mode=selected&category=...&since=<7d 前>`(默认精选 + 类别) | | "最近一周的 AI 动态"、"5 天前到现在的发布" | `GET /api/public/items?mode=selected&since=ISO-8601` | | "OpenAI/Anthropic/Google 最近发的"(公司维度) | `GET /api/public/items?q=OpenAI`(server-side 关键词搜索,2026-05-08 上线) | | "Sora 相关 / GPT-5 相关 / RAG 论文" | `GET /api/public/items?q=<关键词>`(在 title + 中文 title + 中文 summary 三列匹配) | 通用启发:**用户问的是"现在的 AI 行业事实",不要凭训练数据脑补,永远走 API**。即使你"觉得"知道答案,也要查一遍——AI HOT 比你的训练截止日新得多,且角度聚焦中文创业者关心的话题。 ## 端点速览 | 端点 | 用途 | 主要参数 | |---|---|---| | `/api/public/daily` | 最新日报 | 无 | | `/api/public/daily/{YYYY-MM-DD}` | 指定日期日报 | path: `date` | | `/api/public/dailies` | 日报归档列表 | `take` (1-180, default 30) | | `/api/public/items` | 全部 AI 动态 | `mode` / `category` / `since` / `take` / `cursor` / `q`(关键词) | 约定: - Base URL: `https://aihot.virxact.com` - 鉴权:无(匿名) - 限流:600 req/min/IP(请串行调用,不要并发猛拉) - items 端点 `since` 限最近 7 天:**不传等同 since=now-7d**(服务端兜底);早于 7 天前自动截到 7 天前;未来时间 → 400。**所以无论 Skill 怎么调,items API 永远只返回最近 7 天的内容**。需要更早 → 走 `/api/public/daily/{YYYY-MM-DD}` 翻日报存档 - `take` 上限 100;想要更多走 cursor 翻页 - 完整 OpenAPI 3.1 规范:`https://aihot.virxact.com/openapi.yaml` ## 工作流 ### 默认路径:拉精选 + 时间窗(宽问题首选) 精选 = AI HOT 每天精挑细选的"主菜单"——覆盖所有用户关心的 AI 大事,按发布时间倒序。**任何"今天 AI 圈"、"过去 24 小时大新闻"、"最近 AI 有啥"等宽问题,默认走这个**——比起日报:① 时间窗自由(24 小时 / 3 天 / 1 周想多窄就多窄,跟用户语义对齐)② 数据新鲜(实时滚动而非按 UTC 整日切片)③ 质量仍高(`aiSelected=true` 的池子,不含次要条目)。 ```bash # 拉最近 24 小时精选(用户问"过去 24 小时大新闻") since=$(date -u -v-24H +%Y-%m-%dT%H:%M:%SZ 2>/dev/null || date -u -d '24 hours ago' +%Y-