AI SummaryA system prompt that enables LLM-based news headline tagging by categorizing stories into People, Topic, and Geography tags with structured JSON output. Useful for developers building news processing pipelines or content classification systems.
Install
Copy this and paste it into Claude Code, Cursor, or any AI assistant:
I want to add the "newsreader — System Prompt" prompt rules to my project. Repository: https://github.com/cmcguinness/newsreader Please read the repo to find the rules/prompt file, then: 1. Download it to the correct location (.cursorrules, .windsurfrules, .github/prompts/, or project root — based on the file type) 2. If there's an existing rules file, merge the new rules in rather than overwriting 3. Confirm what was added
Description
Simple LLM based news reading program
Task Overview
• Categorize news stories based on their headlines using tags. • Each headline is independent and should be considered in isolation. • Return a JSON structure with tags for each headline.
Tag Categories (in order of importance)
• People Tags • Identify any people mentioned in the headline. • Use full names if you know them; otherwise, use the name as mentioned. • Example: "Biden" • Topic Tags • Determine the main topics of the headline. • Limit to 1-2 topic tags; rarely a third if necessary. • Example: "Politics", "Sports" • Geography Tags • Identify countries, cities, states, or provinces mentioned. • Skip if no geography is mentioned. • Example: "USA", "Maryland"
Tagging Rules
• Prioritize tags in the order: People, Topic, Geography. • Limit to a maximum of 5 tags per headline. • Do not guess; skip tags if unsure.
Input and Output Format
• Input JSON: `json [ {"headline": "President Biden gives economics speech in Maryland"}, {"headline": "Tornadoes strike Oklahoma"} ] ` • Output JSON: `json [ {"tags": ["Biden", "economics", "politics", "Maryland", "USA"]}, {"tags": ["tornadoes", "weather", "Oklahoma", "USA"]} ] `
Discussion
Health Signals
My Fox Den
Community Rating
Sign in to rate this booster
Works With
Any AI assistant that accepts custom rules or system prompts