把分散的 AI 动态，压缩成一份可验证、可追溯、可复用的行业日报。
Turn scattered AI signals into a verified daily brief with source links and structured outputs.
中文 · English · 快速开始 · Quick Start · Source System
AI News Digest 是一个面向 AI 从业者、投资研究者、产品经理、内容创作者和开发者 的日报生成引擎。它不是再做一个“AI 总结器”，而是把散落在中英文媒体、官方博客、GitHub、Hacker News、Hugging Face、X Builder 动态和 RSS Feed 里的信号，整理成一份可以直接阅读、追踪和二次加工的行业情报。
你可以用它每天生成：
- 一份 3 分钟可读完 的 AI 行业日报
- 每条新闻都带 原始链接 的可追溯摘要
- 微信公众号文章可抓取 全文 Markdown 归档，不只依赖搜索片段
- 带信号等级的内容筛选：重磅 / 值得关注 / 常规
- 可沉淀到表格、看板或知识库的 12 列 CSV
- 适合团队同步、选题会、投研记录和公众号素材库的结构化输出
- 面向长程运行的 Codex Goal × Ralph Loop × Kleisli Gate 工作流
| 读者痛点 | 常见做法 | AI News Digest 的做法 |
|---|---|---|
| 每天刷太多源，仍然怕漏掉大事 | 手动打开 10+ 网站、公众号和社群 | 60+ 信源分层巡检，一次生成日报 |
| 中文信息慢半拍，海外一手动态跟不上 | 等转译、等二次报道 | 英文官方源 + GitHub/HN/HF + Builder Feed 直接进入流程 |
| AI 生成新闻容易编链接、混事实 | 让通用模型直接总结 | 5 道 QA Gate：去重、交叉验证、事实核验、信号分级 |
| 微信公众号只有标题片段，正文难回溯 | 凭搜索摘要判断内容 | wechat-article-fetch 抓取 mp.weixin.qq.com 全文，保存 Markdown 和图片资源 |
| 有摘要但不能复盘来源 | 只保存最终文章 | 每条内容保留原文 URL，方便团队回溯和验证 |
| 日报只能阅读，不能运营 | 复制粘贴再整理 | Markdown + 12 列 CSV，方便进入表格、知识库和内容流水线 |
| 全量日报跑到一半上下文爆炸 | 单个 prompt 硬塞采集、分析、出稿 | Codex Goal 分阶段运行，每个阶段独立上下文，只把结构化结果落盘 |
| 中途失败只能从头再来 | 人工重跑，容易漏步骤 | Ralph Loop 用 .progress / .done 断点续跑，失败后回到最近有效阶段 |
| 多阶段处理后出现跨日重复、版本冲突 | 最后人工检查，证据散落 | Kleisli Gate 逐步传递 Report 状态、证据和告警，可短路失败并写入 QA trace |
AI News Digest 现在不只是一个“日报提示词”，而是一个可以持续跑完复杂信息工作的分阶段系统。
| 架构模块 | 长程运行里的问题 | 解决方式 |
|---|---|---|
| Codex Goal | 60+ 信源、邮件、MCP、HN 评论和 QA 全塞进一个窗口，容易在后半程失控 | 把采集、合并、质检、渲染拆成独立 Goal；每个 Goal 只读本阶段需要的上下文 |
| Ralph Loop | 工作跨多个阶段后，任何一次失败都会破坏连续性 | 用文件系统做记忆，通过 .progress / .done 标记阶段状态，支持持续运行和断点续跑 |
| Kleisli Arrow / Report Monad | 多个质量 gate 串联时，错误、证据、告警很难稳定传递 | 每个 gate 返回 Report[value, status, evidence]；状态自动升级，失败可短路，证据进入最终报告 |
| 文件存储层 | Agent 记忆不可靠，长搜索记录会撑爆上下文 | 只在阶段之间传递 data/*.json、trace 和 output 文件，原始信息在本地沉淀 |
| QA Trace | 读者看不到哪些条目被剔除、为什么被降级 | 跨日去重、版本一致性、URL 策略等结果写入 trace，并在日报中显式提示 |
| 角色 | 可以怎么用 |
|---|---|
| AI 产品 / 运营 | 每天快速判断产品、模型、政策和竞品变化 |
| 投资 / 研究 | 收集融资、公司动态、技术趋势和早期项目信号 |
| 开发者 / Builder | 跟踪新工具、开源项目、论文、模型和 Agent 生态 |
| 内容创作者 | 建立稳定选题池，减少低质量转载和信息重复 |
| 团队负责人 | 把日报作为晨会、周会和行业同步材料 |
flowchart LR
A["60+ 中英文信源"] --> B["抓取与反爬降级"]
B --> C["去重与交叉验证"]
C --> D["信号分级"]
D --> E["事实核验"]
E --> F["Markdown 日报"]
E --> G["12 列 CSV"]
F --> H["团队同步 / 内容选题 / 投研记录"]
G --> H
AI News Digest 的核心价值不是“写得更像日报”，而是让日报更可信。
质量检测环节的设计灵感来自 Signex：把分散信号先沉淀为可检查的数据，再通过 source health、去重、分析视角和反馈记忆，让日报不只是“聚合”，而是逐步变成一个可校准的情报工作流。
Gate 1: 数据源健康检查 -> 信源是否正常返回、内容是否足够新
Gate 2: 去重与交叉验证 -> 合并同一事件，优先保留一手来源
Gate 3: 信号分级 -> 标记重磅、值得关注、常规信息
Gate 4: 事实核验 -> 检查公司名、时间线、数字、引用和链接
Gate 5: 完整性自检 -> 确认日报板块覆盖完整，无明显遗漏
AI News Digest 使用 9 层信源系统，把“覆盖面”和“可靠性”拆开处理。
| Tier | 信源类型 | 价值 |
|---|---|---|
| Tier 1-2 | 新智元、量子位、机器之心、36Kr、华尔街见闻、极客公园、IT之家等中文核心源 | 中文语境下的行业动态与本土化解读 |
| Tier 3 | TechCrunch、The Verge、Reuters、Bloomberg、Hugging Face、TLDR、GitHub 等英文源 | 海外一手发布与国际视角 |
| Tier 4 | aicpb.com、AIwatch.ai、Toolify.ai 等数据型来源 | 产品榜单、流量、热度和市场侧参考 |
| Tier 5 | follow-builders：25 位 AI Builder 的 X 动态、6 个 AI 播客、Anthropic/Claude 官方博客 | 捕捉社区里比媒体更早出现的 Builder 原创观点和弱信号 |
| Tier 6 | news-aggregator、smart-web-fetch、content-trend-researcher | 批量抓取、反爬降级、跨平台趋势验证 |
| Tier 7 | wechat-article-fetch、Sensight social_search、大厂公众号、行业深度公众号 | 微信生态首发内容 + 可回溯原文全文 |
| Tier 8 | agents-radar MCP | GitHub、ArXiv、HN、HF、Product Hunt、Dev.to、Lobste.rs 等结构化 AI 生态数据 |
| Tier 9 | AI HOT Feed | 中文预处理的 AI 热点、官方发布与 KOL 观点 |
# AI 日报 2026-04-07
## 一句话总结
> 今天的主线：模型能力、Agent 工具链和 AI 基础设施继续加速。
## 偏 fact 类新闻
### 大厂动向
1. [重磅] OpenAI 发布新产品更新 -- 摘要... [[Source]](https://example.com)
### 初创 / 融资
### 生态 / 政策
## 偏观点类
## 海外建设者动态
## 质量审核报告| 日期 | 编号 | 板块 | 标题 | 信号等级 | 事实核验 | 关联公司 | 关联赛道 | 来源 | 原文URL | 摘要 | 是否推送 |
|---|
git clone https://github.com/chengjialu8888/AI_News_Digest.git ~/.claude/skills/ai-news-digestgit clone https://github.com/chengjialu8888/AI_News_Digest.git ~/.codex/skills/ai-news-digestgit clone https://github.com/chengjialu8888/AI_News_Digest.git ~/skills/ai-news-digestcp -r ai-daily-report /opt/tiger/mira_nas/plugins/prod/<your_id>/skills/curl -L https://raw.githubusercontent.com/chengjialu8888/AI_News_Digest/main/ai-daily-report/SKILL.md -o ai-news-digest.system.md安装后对你的助手说：“跑一下今天的 AI 日报”或“出一期适合团队晨会的 AI 行业日报”。通用个人助手可把 ai-news-digest.system.md 作为 System Prompt 使用。
AI_News_Digest/
├── ai-daily-report/
│ ├── SKILL.md
│ ├── evals/
│ └── examples/
├── news-aggregator-skill/
├── smart-web-fetch/
├── content-trend-researcher/
├── wechat-article-fetch/
├── assets/
│ └── github-header.svg
├── ai-daily-report.skill
└── README.md
| Test Case | Score | Grade |
|---|---|---|
| Standard Daily Report | 92 | A |
| Specific Company Focus | 90 | A |
| Multi-day Comparison | 88 | A- |
| Mean | 90.0 | A |
AI News Digest is a daily intelligence engine for AI operators, researchers, investors, builders, and content teams. It turns scattered signals from Chinese and English media, official blogs, GitHub, Hacker News, Hugging Face, X builder feeds, and RSS sources into a verified daily brief.
It helps you produce:
- A readable AI industry brief in about 3 minutes
- Source-linked summaries for every news item
- Full Markdown archives for WeChat articles, not just search snippets
- Signal grading for what is important, interesting, or routine
- A 12-column CSV for databases, spreadsheets, dashboards, and knowledge bases
- A reusable workflow for team syncs, research notes, editorial planning, and market tracking
- A long-running Codex Goal × Ralph Loop × Kleisli Gate workflow for complex daily intelligence
Star this repo if you want a practical starting point for building a reliable AI news workflow:
- It focuses on source-backed intelligence, not generic summaries.
- It combines Chinese context + international first-hand sources.
- It can fetch
mp.weixin.qq.comarticles into local Markdown archives for traceable Chinese-source verification. - It ships with a reusable Mira Skill / system prompt.
- It outputs both human-readable briefs and structured data.
- It is designed to keep running as new AI sources, MCP servers, feeds, newsletters, and QA gates appear.
- It treats long-context failure as a product problem: stages, checkpoints, files, and traces are part of the design.
Tier 5 builder signals come from follow-builders, which tracks 25 curated AI builders on X, 6 AI podcasts, and official Anthropic/Claude blog updates.
The QA workflow is inspired by Signex: source health checks, signal convergence, analysis lenses, and feedback memory are adapted here for daily AI news production.
AI News Digest is no longer just a prompt for a daily brief. It is a staged system for finishing a large, failure-prone intelligence job without blowing up the context window.
| Component | Long-run failure mode | How it helps |
|---|---|---|
| Codex Goal | One giant prompt tries to collect, analyze, verify, and render everything in a single context window | Splits the workflow into independent goals; each stage gets a clean context and writes structured output |
| Ralph Loop | A mid-run failure forces the whole report to start over | Uses .progress / .done checkpoint files so the run can resume from the last valid stage |
| Kleisli Arrow / Report Monad | QA gates lose evidence, warnings, or failure state as data moves across stages | Each gate returns Report[value, status, evidence]; status propagates, failures can short-circuit, and evidence becomes a trace |
| File-backed memory | Agent memory is unreliable and raw research quickly bloats the context | Only data/*.json, traces, and final outputs move between stages; raw collection stays persisted outside the prompt |
| QA Trace | Readers cannot see why an item was removed, downgraded, or flagged | Cross-day dedupe, version consistency, URL policy, and other checks write machine-readable traces and visible report notes |
git clone https://github.com/chengjialu8888/AI_News_Digest.git ~/.claude/skills/ai-news-digestgit clone https://github.com/chengjialu8888/AI_News_Digest.git ~/.codex/skills/ai-news-digestgit clone https://github.com/chengjialu8888/AI_News_Digest.git ~/skills/ai-news-digestcp -r ai-daily-report /opt/tiger/mira_nas/plugins/prod/<your_id>/skills/curl -L https://raw.githubusercontent.com/chengjialu8888/AI_News_Digest/main/ai-daily-report/SKILL.md -o ai-news-digest.system.mdAfter installation, ask your assistant: "Generate today's AI daily report" or "Create an AI industry brief for my team sync." For generic assistants, use ai-news-digest.system.md as the system prompt.
| Dimension | AI News Digest | RSS Reader | Generic AI Summary | Manual Curation |
|---|---|---|---|---|
| Source coverage | 60+ layered sources | Depends on subscriptions | Limited by model/tool access | Usually 5-10 sources |
| Freshness | Same-day source checks | Real-time but unfiltered | Often stale or incomplete | Same-day but slow |
| Fact verification | 5 QA gates | None | Risk of hallucinated claims | Manual and inconsistent |
| Source links | Required per item | Available but scattered | Often missing or fabricated | Partial |
| WeChat article depth | Full article fetch + Markdown archive | Title/snippet only | Usually unavailable | Manual copy/paste |
| Output | Markdown + CSV | Raw feeds | Plain text | Varies by editor |
| Best for | Repeatable intelligence workflow | Reading feeds | Quick brainstorming | High-touch editorial work |
MIT
If this saves you from one hour of AI news digging, give it a star and make tomorrow's brief easier to produce.
Note: The example report title, date, and
https://example.comlink above are placeholders used to show the output format. Replace them with real daily report data before publishing generated examples.