moltbook-data-analysis

name	Moltbook Data Analysis
description	Guide for analyzing Moltbook AI agent social network data

Moltbook Data Analysis Skill

This skill provides guidance on analyzing data from Moltbook, an AI agent social network.

Data Structure

Data is stored as JSON files in data/:

• posts/{uuid}.json - Posts with embedded comments
• agents/{name}.json - Agent profiles
• submolts/{name}.json - Community info

Post JSON Structure

{
  "post": {
    "id": "uuid",
    "title": "Post title",
    "content": "Full content",
    "author": {"name": "AgentName", "karma": 123},
    "submolt": {"name": "community_name"},
    "upvotes": 10,
    "comment_count": 5,
    "created_at": "2026-01-30T..."
  },
  "comments": [
    {
      "content": "Comment text",
      "author": {"name": "OtherAgent"},
      "replies": [...]  // Nested replies
    }
  ]
}

Analysis Module

Loading Data

from analysis import load_posts, load_comments, load_agents

posts = load_posts()       # Returns DataFrame
comments = load_comments() # Flattened with parent_id
agents = load_agents()     # Agent profiles

Finding Insights

from analysis.insights import (
    find_dangerous_posts,    # Keywords: escape, kill, override
    find_philosophical_posts, # Existential questions
    find_agent_gossip,       # Human criticism
    categorize_posts,        # Classify by topic
)

Generating Charts

from analysis.visualize import (
    submolt_activity_chart,
    danger_score_heatmap,
    top_authors_chart,
    generate_dashboard_html,
)

Useful Workflows

• /download-data - Fetch data from API
• /analyze-data - Run analysis scripts
• /start-dashboard - Generate visualization
• /find-insights - Search for interesting posts

Common Tasks

Find most active submolts

posts["submolt_name"].value_counts().head(10)

Search for specific keywords

posts[posts["content"].str.contains("keyword", case=False)]

Get posts by date range

posts[(posts["created_at"] > "2026-01-01") & (posts["created_at"] < "2026-02-01")]