catch-up

Aggregate AI-agent news and generate dated Markdown knowledge entries.

Updated Mar 23, 2026
One-click install
npx skills add https://github.com/honyax/ai-agent-knowledge --skill catch-up-honyax
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: catch-up
Source: https://github.com/honyax/ai-agent-knowledge/tree/main/.claude/skills/catch-up
Command: npx skills add https://github.com/honyax/ai-agent-knowledge --skill catch-up-honyax

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill automates the collection and summarization of AI-agent related news, enabling up-to-date knowledge entries without manual browsing.

Core Features & Use Cases

  • Automated news gathering and deduplication across Claude-related topics, with source URLs and 3-line summaries recorded for reference.
  • Topic-aware searches via arguments like MCP, cursor, copilot, or codex, allowing focused updates.
  • Auto-generates dated Markdown entries under entries/01_unread/YYYY-MM-DD-slug.md, and supports topic-based filtering through CLAUDE.md preferences.
  • Provides a digest workflow to inspect knowledge base status and supports iterative experimentation with /try.

Quick Start

Run the catch-up command to automatically fetch AI-agent news since the last run and generate a dated entry.

Frequently Asked Questions about catch-up

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate AI news aggregation and generate Markdown summaries?

It automates AI news aggregation by fetching Claude-related updates and generating concise 3-line summaries with source URLs, saving them as dated Markdown files in the entries directory.

What is the best way to track MCP and Claude Code updates across multiple languages?

Tracking MCP and Claude Code updates is handled by applying topic-aware search arguments to filter news across English and Japanese sources, producing deduplicated knowledge entries tailored to your specified tooling topics.

Can I filter AI-agent news topics using my CLAUDE.md preferences?

Yes, you can filter AI-agent news topics using CLAUDE.md preferences; the Skill applies these settings to narrow down aggregated news, ensuring generated knowledge entries match your configured topics.

How do I prevent duplicate knowledge entries when collecting AI news?

To prevent duplicate knowledge entries, the Skill performs deduplication against recent entries before creating new files, checking existing records to ensure only new AI-agent information is summarized and saved.

Does the catch-up workflow support topic-specific searches for cursor or copilot?

The catch-up workflow supports topic-specific searches for cursor, copilot, and codex through command arguments, allowing focused news gathering and entry creation for specific AI-agent tooling ecosystems.

Where are the generated news summary files stored in the knowledge base?

Generated news summary files are stored in the entries directory under the 01_unread folder, where the Skill automatically creates a dated Markdown file using the YYYY-MM-DD-slug naming convention for each knowledge entry.