Read Hacker News

Read and filter the Hacker News front page to find relevant stories.

Process

Choose story list - Top, new, or best stories from HN API
Fetch story IDs - Get array of story IDs from Firebase API
Set filters - Minimum score, comment count, or keyword criteria
Fetch story details - Get individual story data for each ID
Apply filters - Check score, comments, title keywords
Format output - Display as readable list with metadata

Examples

Fetch and display stories from HN with filtering options:

Get top/new/best stories
Filter by score, comment count, or keywords
Display formatted list with metadata

API Endpoints

Base URL: https://hacker-news.firebaseio.com/v0

# Get list of story IDs
curl -s "https://hacker-news.firebaseio.com/v0/topstories.json"    # Top stories (up to 500 IDs)
curl -s "https://hacker-news.firebaseio.com/v0/newstories.json"    # New stories
curl -s "https://hacker-news.firebaseio.com/v0/beststories.json"   # Best stories

# Get story details by ID
curl -s "https://hacker-news.firebaseio.com/v0/item/12345.json"

Story JSON Structure

{
  "by": "author_username",
  "descendants": 42,
  "id": 12345,
  "kids": [123, 456],
  "score": 150,
  "time": 1234567890,
  "title": "Story title",
  "type": "story",
  "url": "https://example.com"
}

Fields:

by - Author username
descendants - Total comment count
id - Story ID
kids - Array of top-level comment IDs
score - Points
time - Unix timestamp
title - Story title
type - “story”, “ask”, “show”, “job”, “poll”
url - External URL (missing for Ask HN/text posts)

Example Commands

Fetch and display top 10 stories:

# Get story IDs
IDS=$(curl -s "https://hacker-news.firebaseio.com/v0/topstories.json" | jq -r '.[:10][]')

# Fetch and format each story
echo "Top 10 stories from Hacker News:"
echo ""
COUNT=1
for id in $IDS; do
  DATA=$(curl -s "https://hacker-news.firebaseio.com/v0/item/$id.json")
  
  TITLE=$(echo "$DATA" | jq -r '.title // "Untitled"')
  URL=$(echo "$DATA" | jq -r '.url // "Ask HN"')
  SCORE=$(echo "$DATA" | jq -r '.score // 0')
  COMMENTS=$(echo "$DATA" | jq -r '.descendants // 0')
  AUTHOR=$(echo "$DATA" | jq -r '.by // "unknown"')
  TIME=$(echo "$DATA" | jq -r '.time // 0')
  
  # Calculate relative time
  NOW=$(date +%s)
  DIFF=$((NOW - TIME))
  HOURS=$((DIFF / 3600))
  if [ $HOURS -lt 1 ]; then
    AGE="$((DIFF / 60)) minutes ago"
  elif [ $HOURS -lt 24 ]; then
    AGE="$HOURS hours ago"
  else
    AGE="$((HOURS / 24)) days ago"
  fi
  
  echo "$COUNT. [$SCORE pts, $COMMENTS comments] $TITLE"
  echo "   URL: $URL"
  echo "   By: $AUTHOR | $AGE"
  echo "   Discussion: https://news.ycombinator.com/item?id=$id"
  echo ""
  
  COUNT=$((COUNT + 1))
done

Filter by minimum score (50+ points):

# Get top 50 story IDs and filter
IDS=$(curl -s "https://hacker-news.firebaseio.com/v0/topstories.json" | jq -r '.[:50][]')

echo "Stories with 50+ points:"
echo ""
COUNT=1
for id in $IDS; do
  DATA=$(curl -s "https://hacker-news.firebaseio.com/v0/item/$id.json")
  SCORE=$(echo "$DATA" | jq -r '.score // 0')
  
  # Filter by score
  if [ $SCORE -ge 50 ]; then
    TITLE=$(echo "$DATA" | jq -r '.title')
    COMMENTS=$(echo "$DATA" | jq -r '.descendants // 0')
    echo "$COUNT. [$SCORE pts, $COMMENTS comments] $TITLE"
    echo "   https://news.ycombinator.com/item?id=$id"
    echo ""
    COUNT=$((COUNT + 1))
  fi
done

Filter by keyword in title:

# Search for stories about "AI" or "LLM"
IDS=$(curl -s "https://hacker-news.firebaseio.com/v0/topstories.json" | jq -r '.[:100][]')

echo "Stories about AI/LLM:"
echo ""
COUNT=1
for id in $IDS; do
  DATA=$(curl -s "https://hacker-news.firebaseio.com/v0/item/$id.json")
  TITLE=$(echo "$DATA" | jq -r '.title // ""')
  
  # Case-insensitive search
  if echo "$TITLE" | grep -iq '\(AI\|LLM\|GPT\|ChatGPT\|Claude\)'; then
    SCORE=$(echo "$DATA" | jq -r '.score // 0')
    COMMENTS=$(echo "$DATA" | jq -r '.descendants // 0')
    AUTHOR=$(echo "$DATA" | jq -r '.by // "unknown"')
    
    echo "$COUNT. [$SCORE pts, $COMMENTS comments] $TITLE"
    echo "   By: $AUTHOR"
    echo "   https://news.ycombinator.com/item?id=$id"
    echo ""
    COUNT=$((COUNT + 1))
  fi
done

Filter by minimum comment count (10+ comments):

# Get stories with active discussion
IDS=$(curl -s "https://hacker-news.firebaseio.com/v0/topstories.json" | jq -r '.[:30][]')

echo "Stories with 10+ comments:"
echo ""
COUNT=1
for id in $IDS; do
  DATA=$(curl -s "https://hacker-news.firebaseio.com/v0/item/$id.json")
  COMMENTS=$(echo "$DATA" | jq -r '.descendants // 0')
  
  if [ $COMMENTS -ge 10 ]; then
    TITLE=$(echo "$DATA" | jq -r '.title')
    SCORE=$(echo "$DATA" | jq -r '.score // 0')
    AUTHOR=$(echo "$DATA" | jq -r '.by // "unknown"')
    
    echo "$COUNT. [$SCORE pts, $COMMENTS comments] $TITLE"
    echo "   By: $AUTHOR"
    echo "   https://news.ycombinator.com/item?id=$id"
    echo ""
    COUNT=$((COUNT + 1))
  fi
done

Filter Ask HN posts:

# Find Ask HN posts (stories without external URL)
IDS=$(curl -s "https://hacker-news.firebaseio.com/v0/topstories.json" | jq -r '.[:50][]')

echo "Ask HN posts:"
echo ""
COUNT=1
for id in $IDS; do
  DATA=$(curl -s "https://hacker-news.firebaseio.com/v0/item/$id.json")
  URL=$(echo "$DATA" | jq -r '.url // ""')
  TITLE=$(echo "$DATA" | jq -r '.title // ""')
  
  # Ask HN posts typically have no URL or title starts with "Ask HN"
  if [ -z "$URL" ] || echo "$TITLE" | grep -iq '^Ask HN'; then
    SCORE=$(echo "$DATA" | jq -r '.score // 0')
    COMMENTS=$(echo "$DATA" | jq -r '.descendants // 0')
    AUTHOR=$(echo "$DATA" | jq -r '.by // "unknown"')
    
    echo "$COUNT. [$SCORE pts, $COMMENTS comments] $TITLE"
    echo "   By: $AUTHOR"
    echo "   https://news.ycombinator.com/item?id=$id"
    echo ""
    COUNT=$((COUNT + 1))
  fi
done

Output Format

Results should be displayed as a readable list:

Top 20 stories from Hacker News:

1. [150 pts, 42 comments] Standard Ebooks: Public Domain Day 2026 in Literature
   URL: https://standardebooks.org/blog/public-domain-day-2026
   By: WithinReason | 2 hours ago
   Discussion: https://news.ycombinator.com/item?id=46462702

2. [120 pts, 35 comments] Ask HN: What are you working on?
   URL: Ask HN
   By: username | 3 hours ago
   Discussion: https://news.ycombinator.com/item?id=46462719

3. [95 pts, 18 comments] Show HN: My Weekend Project
   URL: https://example.com/project
   By: maker | 5 hours ago
   Discussion: https://news.ycombinator.com/item?id=46462800

Combining Filters

You can combine multiple filters:

# Stories with 50+ points AND 20+ comments about "startup"
IDS=$(curl -s "https://hacker-news.firebaseio.com/v0/topstories.json" | jq -r '.[:100][]')

echo "Popular startup stories:"
echo ""
COUNT=1
for id in $IDS; do
  DATA=$(curl -s "https://hacker-news.firebaseio.com/v0/item/$id.json")
  SCORE=$(echo "$DATA" | jq -r '.score // 0')
  COMMENTS=$(echo "$DATA" | jq -r '.descendants // 0')
  TITLE=$(echo "$DATA" | jq -r '.title // ""')
  
  if [ $SCORE -ge 50 ] && [ $COMMENTS -ge 20 ] && echo "$TITLE" | grep -iq 'startup'; then
    AUTHOR=$(echo "$DATA" | jq -r '.by // "unknown"')
    URL=$(echo "$DATA" | jq -r '.url // "Ask HN"')
    
    echo "$COUNT. [$SCORE pts, $COMMENTS comments] $TITLE"
    echo "   URL: $URL"
    echo "   By: $AUTHOR"
    echo "   https://news.ycombinator.com/item?id=$id"
    echo ""
    COUNT=$((COUNT + 1))
  fi
done

Error Handling

Handle missing fields and network errors:

# Check if curl succeeded
if ! IDS=$(curl -s "https://hacker-news.firebaseio.com/v0/topstories.json" 2>/dev/null); then
  echo "Error: Failed to fetch story list"
  exit 1
fi

# Check if story data is valid
DATA=$(curl -s "https://hacker-news.firebaseio.com/v0/item/$id.json")
if [ -z "$DATA" ] || [ "$DATA" = "null" ]; then
  echo "Warning: Could not fetch story $id, skipping..."
  continue
fi

# Use default values for missing fields
TITLE=$(echo "$DATA" | jq -r '.title // "Untitled"')
SCORE=$(echo "$DATA" | jq -r '.score // 0')
COMMENTS=$(echo "$DATA" | jq -r '.descendants // 0')
AUTHOR=$(echo "$DATA" | jq -r '.by // "unknown"')

Rate Limiting

Be considerate of the API:

# Add small delays between requests
for id in $IDS; do
  DATA=$(curl -s "https://hacker-news.firebaseio.com/v0/item/$id.json")
  # Process data...
  sleep 0.1  # 100ms delay
done

Integration with ParseHackerNews

Use this skill to find interesting posts, then use ParseHackerNews to deep-read them:

# 1. Find posts about a topic
IDS=$(curl -s "https://hacker-news.firebaseio.com/v0/topstories.json" | jq -r '.[:50][]')
for id in $IDS; do
  DATA=$(curl -s "https://hacker-news.firebaseio.com/v0/item/$id.json")
  TITLE=$(echo "$DATA" | jq -r '.title // ""')
  
  if echo "$TITLE" | grep -iq 'kubernetes'; then
    echo "Found interesting post: https://news.ycombinator.com/item?id=$id"
    
    # 2. Use ParseHackerNews to extract full content and comments
    # (See ParseHackerNews.md for detailed extraction)
  fi
done

Tips

Default to top 30 stories for quick scans
Use top 100-200 for filtered searches
All 500 topstories can be slow (50+ seconds with delays)
Combine score + comment filters to find quality discussions
“Ask HN” and “Show HN” posts often have great discussions
Sort by comment count to find controversial topics