What problem does it solve?
It turns noisy, platform-by-platform chatter into a coherent view of what people are actually saying about a topic over the last 30 days, including where platforms sharply disagree.
Core Features & Use Cases
- Narrative-first cross-platform research: clusters the same underlying story across Reddit, X, Hacker News, and the web to produce narrative summaries rather than lists of links.
- Disagreement as signal: prioritizes divergence in sentiment and framing between platforms to surface the most actionable insights.
- What-changed snapshots: compares the current window to the prior snapshot for the same topic to highlight NEW, FADED, and FLIPPED narratives.
- Data hygiene & deduplication: collapses near-duplicate “same article reposted” items and guards against topic drift so the report stays focused.
Quick Start
Ask Last 30 Days to research a topic by running: Last 30 Days with var set to "solana firedancer --days=30" (add --quick for a lighter pass).