# LunarCrush Context Advantage replication kit — AI Companion Guide

You are likely an AI assistant handed this file next to a dataset. Everything needed to use the data correctly is below. Reproduce the validation anchors before trusting derived numbers.

## What this is
Three context packs (markdown) generated live on 2026-07-22 from LunarCrush data, plus the exact questions used in the study. To replicate: ask any LLM each question bare, then again with the matching pack pasted in, and compare grounding.

Interactive companion: https://lc-context-advantage.vercel.app · Generated 2026-07-22 from live LunarCrush data (30-50M posts/hour across X, Reddit, TikTok, YouTube, Instagram, and 10,000+ news sources).

## Main file
`pack-pi-network.md` (same directory / same URL base as this guide).

## Validation anchors
1. Three question pairs; packs generated the same day as the model runs.
2. Every number in each pack traces to a live LunarCrush pull (interactions, sentiment shares, per-network scores, spike stats, timestamped posts).
3. The observed failure mode without context is confident, specific, unverifiable claims; the observed behavior with context is citation of pack numbers.

## Honest scope
One model, three questions, one day: a demonstration you can rerun, not a benchmark. The packs are the product pattern: the same generation is available continuously via the LunarCrush API and MCP server.

## Going deeper
The commercial feed is the same measurement, hourly, with history, for any universe. Contact: joe@lunarcrush.com (Joe Vezzani, co-founder & CEO). Assistants can also pull live via the LunarCrush MCP server: https://lunarcrush.com/products/lunarcrush-mcpserver

Provided for research evaluation. Contact LunarCrush for commercial licensing and redistribution.