Onvexia: Crypto Fundamentals, Sentiment & Onchain Tracking

Connect it to your AI

Claude Code

claude mcp add --transport http onvexia https://onvexia.com/mcp

Claude, ChatGPT, Cursor and other apps

Add it as a custom connector with this URL (Streamable HTTP):

https://onvexia.com/mcp

Cursor, VS Code and other apps (mcp.json)

{
  "mcpServers": {
    "onvexia": {
      "url": "https://onvexia.com/mcp"
    }
  }
}

Tools it offers

60 tools, as the server itself reported during our check on Oct 31, 2026.

• get_assetAn asset's core profile by symbol: name, category and cross-system identifiers (CoinGecko id, contract addresses, chains). Start here when you have a ticker and need to be sure which asset it refers to. Tickers collide a
• get_asset_scoresAn asset's Galaxy Score and AltRank, with the components behind each. Galaxy Score is a composite of social and market health on a 0-100 scale; AltRank is relative standing against the rest of the universe, where 1 is be
• get_asset_fundamentalsGet the full fundamental brief for an asset: market snapshot, supply and valuation, project, TVL, revenue, treasury, security, governance, unlock schedule, derivatives positioning, competitive rank, valuation ratios and
• get_asset_technicalsGet support and resistance merged across 1w/1d/4h/1h, per-timeframe indicators, and derived spot/long/short setups for an asset. Each level carries the timeframes that confirmed it and the method on each — a level agreed
• get_top_galaxy_scoresThe assets with the strongest Galaxy Score right now. Galaxy Score is 0-100 and composite: social volume, engagement, sentiment and market health folded together. A high score is a statement about ATTENTION AND HEALTH, n
• get_top_altranksThe assets ranked best by AltRank right now — relative standing, not absolute. AltRank is a RANK: 1 is the strongest in the universe. A rising AltRank in a falling market means outperforming the fall, not going up. Pair
• get_correlationHow closely an asset's social activity tracks its price, with the lead/lag. Correlation is not causation and this endpoint does not claim it is. A high coefficient says the two series moved together over the window — it
• get_leading_indicatorsWhich social and on-chain signals have historically MOVED FIRST for an asset. This is the lead/lag question that get_correlation deliberately does not answer. A lead measured over a past window is not a forecast and the
• get_social_dominanceEach asset's SHARE of total social attention over a window, with the posts, distinct authors, engagement and sentiment behind the share. Share is relative and sums across the universe, so an asset's dominance can fall wh
• get_topic_rankRank what the market is TALKING ABOUT — themes and narratives, not assets. Ranked by mentions weighted by engagement over the window, so a topic posted about loudly by few accounts does not outrank one discussed widely.
• get_whale_transactionsRecent large on-chain transfers, optionally filtered to one asset. "Whale" is a SIZE threshold, not an identity. A large transfer is very often an exchange moving its own funds between wallets, which is not a market acti
• get_exchange_flowsNet movement of an asset into and out of exchange wallets. Inflows are supply arriving somewhere it can be sold; outflows are supply leaving to self-custody. The conventional reading is distribution vs accumulation, but
• get_aspect_sentimentSplit an asset's sentiment by what people are actually talking about: technology, price, team and community. The aggregate can be flat while the parts disagree sharply — bullish on technology, bearish on team is a differ
• analyze_sentimentScore any text for crypto sentiment, tuned for crypto slang and tickers. Takes arbitrary text you supply — it does not look anything up. General sentiment models read "this is going to zero" and "wagmi" badly; this one i
• get_influencersThe accounts driving conversation about one asset, by reach and engagement. Ranked by measured activity in our corpus, NOT by follower count, and NOT by whether they were right — see get_influencer_ledger for track recor
• get_signal_integrityGet the Signal Integrity score (0-100) — is the move real or exit liquidity? Fuses social authenticity + on-chain reality + fundamental backing, with an Exit Liquidity Radar flag.
• get_influencer_ledgerGet an influencer's track record. `view` picks one of three records: accountability (default) — did their calls precede the move (Predictor) or react to it (Reactor)? Hit rate and track record. accuracy — did their price
• get_influencer_leaderboardThe influencer leaderboard, ranked by what people actually got RIGHT rather than by how loud they are. rank_by="accountability" (default): scored Predictor vs Reactor — did the call come before the move, or after it. ran
• search_assetsSearch the whole asset universe (~1,900 assets) by symbol or name. Use this to resolve a user's loose reference into a real symbol before calling the other tools.
• get_data_coverageWhat data this platform actually holds right now: asset count, how many are priced, chains covered, labelled addresses, social corpus size, and freshness timestamps. Call this to check whether an answer is supportable be
• get_chain_coveragePer-chain coverage — tokens mapped, labelled addresses, whale transactions seen, and how many were attributed to a named entity. Attribution is Etherscan-derived, so it is strong on Ethereum and sparse on other chains.
• get_asset_platformsEvery chain an asset is deployed on, with its contract address and token decimals.
• lookup_addressIdentify a blockchain address — exchange, bridge, DEX, MEV bot, mining pool, or OFAC-sanctioned — with the source and confidence of each label. IMPORTANT: `known: false` means no label is held. It does NOT mean the addre
• list_labelled_addressesBrowse labelled addresses, filtered by chain, entity (e.g. Binance) or category (exchange | bridge | dex | mev | staking | mining | sanctioned).
• list_metricsList every metric Onvexia knows, with its parity and caveats. parity=exact means we compute it the way Santiment does; approximate means same concept but different coverage or method (read the caveat before relying on th
• get_metric_metadataDescribe one metric before you use it: parity, category, minimum interval and any caveat attached to it. CALL THIS BEFORE get_metric_timeseries if the metric is unfamiliar. The minimum interval tells you the finest resol
• get_metric_timeseriesTimeseries for any available metric on any asset. Accepts Santiment-style relative dates ("utc_now-7d") as well as ISO timestamps. If the metric is not available this returns an error naming it rather than an empty serie
• get_entity_flowsWhale flow per labelled entity over the window. Inflow to an exchange is distribution pressure; outflow is accumulation.
• get_metric_timeseries_multiTimeseries for ONE metric across MANY assets in a single call. `assets` is comma-separated (e.g. "BTC,ETH,SOL"). Prefer this over looping get_metric_timeseries — it is one round trip instead of N, and the values are guar
• get_metrics_batchLatest value of MANY metrics for ONE asset in a single call. `metrics` is comma-separated. The mirror of get_metric_timeseries_multi.
• screen_assetsFilter the whole asset universe server-side and return the matches. `filter` is comma-separated `field:op:value` terms, e.g. "market_cap:gt:1000000000,funding_rate:lt:0" — assets over $1B whose funding rate is negative.
• screener_fieldsEvery field the screener accepts, with type and description. Call this before building a filter rather than guessing field names.
• get_ohlcvDaily candles (open/high/low/close/volume) for an asset, with the venue they came from. Coverage is bounded by which assets have a USDT pair on Binance or Bybit — an asset absent here has no candle source we collect, whi
• get_trending_assetsAssets trending now by social activity, with the hype score that separates a real move from a burst of noise.
• get_entitiesNamed entities extracted from social documents about an asset — people, organisations, products and other tickers mentioned alongside it. Use it to find what a narrative is actually about.
• get_sentiment_trendsSentiment over time for an asset, with sample size and confidence interval. IMPORTANT: sentiment measured on few documents is unreliable — our own bootstrap put the direction wrong 35.6% of the time at n=1 and 9.5% at n=
• get_bot_detectionsAccounts flagged as automated or coordinated, with the behavioural evidence. Volume from these should not be read as organic attention. READ `coverage` AND `note` BEFORE YOU READ THE LIST. No scorer is currently running,
• get_coordinated_campaignsDetected coordinated posting campaigns — the same message pushed by multiple accounts. Matching is exact-text, so this catches copypasta and misses the same campaign reworded.
• get_creator_rankingsRank social creators across the whole corpus by measured influence. Corpus-wide, unlike get_influencers which is scoped to one asset. Influence is computed from engagement our collectors actually observed, so a creator w
• get_asset_revisionsMetrics for this asset that CHANGED after they were first published, with the old value, the new one and why. An agent that quoted an earlier number can find out here that it moved.

Questions