跳到正文
TakoAPI
返回技能列表

feed-diet

由 tkuehnl 提供

Health & Fitness
405下载量由 clawskills.sh 提供
1浏览量
0 个赞登录后即可为这个技能点赞
—暂无评分
你的评分

简介

Audit your information diet across HN and RSS feeds — beautiful reports with category breakdowns, ASCII charts.

描述

该技能的功能

Audits information consumption from Hacker News and RSS feeds. Produces Markdown reports with category breakdowns, diversity scores, and reading recommendations. Supports a digest mode that filters recent content by user-defined goals.

Combines Hacker News history and OPML feed data in one workflow with LLM-powered classification and goal-based digest filtering, without requiring any external dashboard or service.

何时使用

  • Finding out if your HN reading skews too heavily toward news vs technical content
  • Getting a weekly digest filtered to your work on compilers or distributed systems
  • Auditing RSS subscriptions to see which topics dominate your feed
  • Discovering unexpected category patterns in your reading history
  • Building a goal-filtered reading list from the past week's HN content

示例工作流

输入

User asks: Audit my HN reading diet — my username is tosh

智能体
  1. 1

    Fetch recent HN stories for the given username via hn-fetch.sh

  2. 2

    Classify fetched items using classify.sh with LLM or keyword fallback

  3. 3

    Run feed-diet.sh audit --hn tosh --limit 100 to generate the full report

  4. 4

    Present the complete Markdown report with category table and recommendations

输出

Markdown report with category breakdown table, top topics with sample items, diversity score, and personalized reading recommendations

前置要求

  • ANTHROPIC_API_KEY environment variable (optional, improves classification accuracy over keyword fallback)
  • OPENAI_API_KEY environment variable (optional, improves classification accuracy over keyword fallback)

安装

clawhub install tkuehnl/feed-diet

运行 CLI 命令,或把这段提示词粘贴到 Claude Code、Cursor、Codex 或你在用的任意编程智能体里。

链接