Sharpe Terminal vs Kaito

Strategic choice. Kaito's view is that the formula is the moat and gating it behind enterprise pricing both protects the IP and signals quality to institutional clients. Sharpe's view is that publishing the methodology earns trust, lets traders sanity-check the score, and earns citations from researchers and journalists. Both views are defensible — the user's choice depends on whether they value the proprietary signal or the open verifiability more.

Yes, sometimes meaningfully. Kaito's yapper-weighted Twitter/X model can rank a narrative high because a few large-following accounts are posting about it, even without broader engagement growth. Sharpe's multi-signal model can rank the same narrative lower because its Telegram, watchlist, and GitHub signals are flat. Conversely, Sharpe can rank a developer-heavy narrative (e.g. Modular blockchains) high while Kaito ranks it lower because Twitter/X attention is muted.

Layer 1, Layer 2, DeFi, AI Agents, DeFAI, DePIN, DeSci, Gaming, DEX, CEX tokens, Lending, Memes, NFTs, Oracles, Privacy, RWA, Stablecoins, Restaking, Liquid Staking, Modular blockchains, SocialFi, and Intent-based protocols.

Hourly. The cron pulls Twitter/X mention volume and engagement every 4 hours, Telegram membership and watchlist growth hourly, and GitHub commits daily. Each narrative's composite score is recomputed on each pull. The widget timestamp shows exact freshness.

Yes. The /api/mindshare endpoint returns the full 22-narrative leaderboard with all underlying signals (totalSocialScore, mindsharePct, sentimentPct, telegramMembers, watchlistUsers, totalCommits4w, totalGithubStars, delta24h, delta7d). Free tier is 30 req/min, 10,000 req/month.

Sharpe filters obvious bot patterns: engagement-to-follower ratios above 5x (engagement farming), sub-second posting cadence (programmatic accounts), and account-age clustering (coordinated newly-created accounts). Filtered accounts are excluded from the social score calculation. The full filter logic is documented at /methodology/mindshare-score.

Yes. Each narrative on the mindshare tracker can be drilled into a per-token view showing the underlying token-level social scores. Visit /mindshare/{narrative-slug} to see the breakdown.