Menu
Removes extraneous information (source, author, etc.) from link titles using LLM intelligence.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
| name | clean_link_titles |
| description | Removes extraneous information (source, author, etc.) from link titles using LLM intelligence. |
This skill uses an LLM to intelligently clean news link titles by removing unnecessary metadata such as source names, author names, and site branding, while preserving the core headline.
TempLinks.md).[Title](URL), identify and remove extraneous suffixes or prefixes. | SiteName (e.g., | TechNews, | iThome, - MoneyDJ). - Source Name (especially if it looks like a brand).(Source) at the end.【Source】 or [Source] if it's clearly a publisher tag.Author Name: if present at start.Input:
- [NVIDIA stock jumps 5% | CNBC](...)
- [【TechCrunch】OpenAI releases new model](...)
- [Review: iPhone 16 | The Verge](...)
Output:
- [NVIDIA stock jumps 5%](...)
- [OpenAI releases new model](...)
- [Review: iPhone 16](...)
Appends relevant hashtags to journal entries based on content analysis. Intended for LLM execution.
Organizes tagged links from a source file into the daily journal structure.
Removes timestamps and decodes URLs in the specified markdown file.
Checks if the daily journal file exists for a given date. If not, creates it using the 'Daily framework' template.
End-to-end workflow to process daily news digest: ensures journal, cleans links, tags, and organizes them.
Empties the content of a specified file. Use with caution.