Skip to main content

news-radar-ai

本地运行的个人 AI 资讯编辑部:消费 newsbox 采集的原始内容, 经 DeepSeek 初筛 + 多维评分,产出一份 Obsidian Markdown AI 日报。

news-radar-ai 是「加工层」——它不直接抓数据,而是从上游 newsbox 采集层读取原始资讯, 用大模型逐条初筛去噪、打分、写摘要、打标,再把候选交给你(或你的 AI 助手)聚类、精选、 写成日报。

架构

newsbox(采集层,独立包)  →  news-radar-ai(本包,加工层)  →  Obsidian Markdown 日报
   RSS / X / Reddit /            初筛 · 评分 · 摘要 · 打标
   官方一手 / 通用网页            (DeepSeek 批量处理)
  • 上游:newsbox —— 负责全部采集,安装本包时会作为依赖自动拉取。
  • 外部模型:DeepSeek —— 逐条评分用,需要一个 API key。

安装

uv tool install news-radar-ai      # 推荐;或 pipx install news-radar-ai

会自动一并安装上游 newsbox 及其采集依赖。

初始化

# 1. 生成配置文件 ~/.news-radar-ai/config.yaml
news-radar-ai config init

# 2. 填入 DeepSeek API key(到 https://platform.deepseek.com 创建)
echo 'DEEPSEEK_API_KEY=sk-你的key' >> ~/.news-radar-ai/.env

# 3. 设置日报输出目录(Obsidian vault 或任意可写目录)
news-radar-ai config set report.obsidian_vault_path '/path/to/your/vault/'

# 4. 初始化数据库
news-radar-ai db init

# 5. 环境自检
news-radar-ai preflight

preflight 会检查 5 项:newsbox 是否装好、DeepSeek key、输出目录、本地数据库、上游 raw.db。

准备上游数据

本包只加工,不采集。先让 newsbox 抓到数据:

newsbox config init
newsbox sources seed       # 载入预置的数十个 AI 信源(多为标准 RSS,开箱即用)
newsbox fetch --since=7d   # 采集最近 7 天

预置信源以 RSS/Atom 为主,无需额外凭证即可抓取;少数信源(如本地 RSSHub 源、X/Twitter) 需要 Docker 或访问凭证,属可选增强。详见 newsbox 文档。

使用

命令行流水线

news-radar-ai pull --since 2026-01-01T00:00:00          # 从 newsbox 拉数据入库
news-radar-ai filter-score                              # DeepSeek 批量初筛 + 评分
news-radar-ai report-candidates --window-start 2026-01-01T00:00:00 --json   # 取候选

report-candidates 输出打过分的候选 JSON,你可据此聚类、精选、写日报。

配合 AI 助手(推荐)

本包设计为配合 AI 编程助手(如 Claude Code)的编排工作流使用:助手按 「拉数据 → 评分 → 取候选 → 写 Markdown」全程跑完,你只需一句「生成过去 24h 的 AI 资讯报告」。

配置速查

news-radar-ai config show                                  # 查看当前配置(密钥已脱敏)
news-radar-ai config set scoring.featured_threshold 75      # 精选门槛
news-radar-ai config set llm.model deepseek-v4-flash        # 切换 DeepSeek 模型

数据与配置都在 ~/.news-radar-ai/(config.yaml / .env / ai.db)。

依赖

Python ≥ 3.11;运行时依赖 newsbox ≥ 1.0.3、typer、pydantic、httpx、loguru 等(自动安装)。

License

MIT

Metadata

Release files for news-radar-ai 1.0.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for news-radar-ai 1.0.0
File Size Uploaded
news_radar_ai-1.0.0.tar.gz 36.3 kB Details

Built distribution (wheel)

Table of built distributions (wheels) for news-radar-ai 1.0.0
File Interpreter ABI Platform
news_radar_ai-1.0.0-py3-none-any.whl Python 3 none any Details

Total release size: 85.9 kB

Release files / news_radar_ai-1.0.0.tar.gz

Download URL news_radar_ai-1.0.0.tar.gz
Size 36.3 kB
Tags Source
SHA-256 checksum
How to use checksums
b760a3d3e593e8811983ef3e53342821a516b4508e1c698febf7de7c2cfd25c5
BLAKE2b-256 checksum
How to use checksums
2298192827d126b844c23e8e09e75ec2c62128abd71a23ab98da242067a8db5c
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via uv/0.7.20

Release files / news_radar_ai-1.0.0-py3-none-any.whl

Download URL news_radar_ai-1.0.0-py3-none-any.whl
Size 49.6 kB
Tags Python 3
SHA-256 checksum
How to use checksums
6140f36be65f21275b6b686280c0d4808924aa2ca732e218e6eee8c70c077cc8
BLAKE2b-256 checksum
How to use checksums
1c7cf7f444e75fe27d174e73663d1bba2ef1db0952a33816638e1d2331e8f039
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via uv/0.7.20

Release history Release notifications | RSS feed

This release

1.0.0 This release

2 release files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page