Allium MCP

Overview

Allium MCP is an official, API-key-gated MCP server run by Allium Labs at mcp.allium.so. It exposes tools to run and manage Explorer SQL queries, fetch and search schemas, and access Allium's Realtime API, covering enterprise blockchain data infrastructure across 80+ chains. An agent builder plugs it in to pull structured, queryable on-chain data instead of building its own indexing pipeline; requires an Allium API key from app.allium.so.

Tool inventory — 48 tools, probed live 2026-09-27

• search_docs — Search Allium's public documentation by meaning. Use for questions about data schemas, APIs, and query guidance when you do not know the document path. Use browse_docs to navigate known paths.
• browse_docs — Browse Allium's public documentation by path. An empty path lists the root; a directory path lists its contents; a Markdown file path returns the file. Read each directory's `_index.md` to choose a do
• search_schemas — Search schemas or fetch one schema entry by id. `query` runs the chosen `search_mode` against the prebuilt schema index. `id` fetches a single entry by its dotted table id (e.g. `ethereum.public.tran
• search_terminal — Search Allium's published `/terminal/` dashboards. Returns ranked discovery details and optionally rendered dashboard content, without SQL or result rows. Use `include_content=True` with a small `limi
• get_terminal_results — Fetch SQL or latest precomputed result rows for one Terminal dashboard chart. Use search_terminal first to find a dashboard id, then call get_terminal_results without chart_id for the chart manifest.
• run_sql_query — Run an ephemeral, throwaway SQL query on Allium's blockchain data. **Call `get_skill(name="sql-optimization")` before writing or running any SQL.** Use ONLY when the result is not reused — inspectin
• get_query_run_results — Get status and results for an ad hoc or saved Explorer query run. Returns columns, rows, total `row_count`, and `cost_explorer_units` when complete; `row_limit` may return fewer rows than `row_count`.
• create_explorer_query — Create and save a SQL query in the user's Allium Explorer. **Call `get_skill(name="sql-optimization")` before writing or running any SQL.** The DEFAULT path whenever the result will be reused — visu
• get_explorer_query — Retrieve a saved Explorer query by ID. Returns the full SQL, row limit, template parameters, tags, URL, and latest result field metadata when available. Use this to inspect a query before editing or
• update_explorer_query — Update an existing Explorer query. Requires query_id. Only provide fields you want to change - others will be preserved.
• list_explorer_queries — List user's saved Explorer queries with field metadata and tags. Use the tags parameter to filter queries by tag (e.g., to find all queries tagged for a specific dashboard or topic).
• run_explorer_query — Run an existing saved Explorer query by ID. Queues the query for execution and returns a run_id. Use get_query_run_results to wait for results or for a non-blocking status check.
• delete_explorer_query — Permanently delete an Explorer query by ID.
• create_explorer_visual — Create or replace a visual on an existing Allium Explorer query. Visuals are persisted to the caller's account and render on the query page; the returned `url` deep-links to the visual this call wrote
• get_explorer_visual — Read one visual saved on an Allium Explorer query, including its full `spec`. Use this to tweak a field of an existing visual, or to match its style in a new one — get_explorer_query lists a query's
• delete_explorer_visual — Delete one visual from an Allium Explorer query. The query and its other visuals are untouched. Get the `visual_id` from get_explorer_query's `visuals`. Editor permission on the query is required, an
• list_compute_profiles — List the organization's Explorer compute profiles with each profile's org-wide `queued` and `running` query counts. Each profile has its own warehouse and its own queue, shared across the org but isol
• read_dashboard — Inspect a dashboard at any depth. Pass `dashboard_id` alone for a root summary (name, settings, list of pages). Add `page_id` / `section_id` / `element_id` to drill down — each level returns the targ
• search_dashboards — Search the user's Explorer dashboards by name. Returns dashboard IDs, names, URLs, tags, timestamps, and whether the caller created each one, ranked by match quality then recent update. With no argum
• set_dashboard — Upsert or delete any node in a dashboard tree. Call get_skill(name='dashboard-design') first — it covers the build order, element catalog, layout, and design/SQL rules. For an element, `config.type`
• delete_dashboard — Delete an entire dashboard, including every page, section, and element inside it. This is irreversible. Confirm with the user which dashboard before calling, and read_dashboard first if you are unsur
• list_skills — List Allium skill guides the caller can access, with descriptions. Use get_skill for full instructions.
• get_skill — Get a skill's full instructions. If the instructions list reference files, call again with `reference_title` set to the exact listed title to retrieve one.
• get_realtime_supported_chains — List supported chains by realtime API endpoint. Call once per session and reuse the endpoint-to-chains map before token, wallet, holdings, or price requests. An unsupported chain may return an empty r
• get_realtime_token_latest_price — Get latest prices for token chain/address pairs, optionally with liquidity data. The request body must be an array, even for a single token. For chain coverage, edge cases, or response fields, use bro
• get_realtime_token_price_history — Get token price candles for one or more token chain/address pairs over a time range at the requested granularity. Use cursor for the next page. For chain coverage, edge cases, or response fields, use
• get_realtime_token_price_stats — Get price stats for token chain/address pairs, including volume, highs, lows, and price changes. The request body must be an array, even for a single token. For chain coverage, edge cases, or response
• get_realtime_token_price_at_timestamp — Get token prices at a timestamp, using the closest earlier price within the optional staleness tolerance. For chain coverage, edge cases, or response fields, use browse_docs with path 'api/developer/p
• get_realtime_wallet_latest_token_balances — Fetch the latest fungible token balances for wallet addresses across supported chains, optionally including liquidity totals. For chain coverage, edge cases, or response fields, use browse_docs with p
• get_realtime_wallet_historical_token_balances — Fetch paginated historical fungible token balances for wallet addresses over a requested time range. For chain coverage, edge cases, or response fields, use browse_docs with path 'api/developer/wallet
• get_realtime_wallet_positions — Retrieve DeFi positions, including LP and lending positions, for up to 5 wallet chain/address pairs. For chain coverage, edge cases, or response fields, use browse_docs with path 'api/developer/defi-p
• get_realtime_wallet_transactions — Get paginated transaction activity for wallet chain/address pairs, including activities, asset transfers, and labels. Deprecated chains return 410; choose a supported chain before retrying. For chain
• list_realtime_chain_tokens — List chain tokens, including tokens without DEX trades. Those tokens omit price-derived fields rather than reporting zero and rank last under FDV or volume sorting. For Stellar, set chain=stellar; sor
• search_realtime_chain_tokens — Search tokens by name or symbol. When chain is set, tokens that have never traded on a DEX are included with price-derived fields omitted. When chain is omitted, cross-chain search only returns tokens
• get_realtime_tokens_by_chain_address — Get token details by chain and address, including tokens without DEX trades. Those tokens retain on-chain decimals and total_supply but omit price-derived fields rather than reporting zero. A not-foun
• get_realtime_holdings_history — Get historical aggregated USD holdings for one or more addresses. For chain coverage, edge cases, or response fields, use browse_docs with path 'api/developer/holdings/holdings-history.md'.
• get_realtime_holdings_pnl — Get profit and loss for up to 20 wallet chain/address pairs. For chain coverage, edge cases, or response fields, use browse_docs with path 'api/developer/holdings/holdings-pnl.md'.
• get_realtime_holdings_pnl_history — Get historical profit and loss for up to 20 wallet chain/address pairs over a time range. For chain coverage, edge cases, or response fields, use browse_docs with path 'api/developer/holdings/holdings
• get_realtime_holdings_pnl_by_token — Get profit and loss for up to 20 wallet-token chain/address triples. For chain coverage, edge cases, or response fields, use browse_docs with path 'api/developer/holdings/holdings-pnl-by-token.md'.
• get_realtime_holdings_pnl_by_token_history — Get historical profit and loss for up to 20 wallet-token chain/address triples over a time range. For chain coverage, edge cases, or response fields, use browse_docs with path 'api/developer/holdings/
• get_realtime_hyperliquid_info — Query Hyperliquid account, market, or exchange information by request type through Allium. Use the dedicated fills, order history, order status, orderbook, or asset contexts tools for those storage-ba
• get_realtime_hyperliquid_fills — Get Hyperliquid userFills, userFillsByTime, or userTwapSliceFills by wallet address. These request types match Hyperliquid; userFillsByTime has no historical lookback limit. Results are capped at 2000
• get_realtime_hyperliquid_order_history — Get a wallet's Hyperliquid historicalOrders. Omit times for recent orders. Historical lookback is unrestricted, but each startTime-endTime window must span at most 30 days; paginate longer ranges with
• get_realtime_hyperliquid_order_status — Get a Hyperliquid order status by wallet and numeric order ID or client order ID. Unknown orders return Hyperliquid's unknownOid status. For chain coverage, edge cases, or response fields, use browse_
• get_realtime_hyperliquid_orderbook_snapshot — Get complete orderbook snapshot for all pairs. Compression Required. This endpoint requires brotli compression for efficient data transfer. For chain coverage, edge cases, or response fields, use brow
• get_realtime_hyperliquid_asset_contexts — Get current or historical Hyperliquid metaAndAssetCtxs snapshots with prices, funding, and open interest. The latest snapshot lags the exchange by about 2-3 minutes. History requires both times and sp
• list_realtime_crosschain_assets — List cross-chain assets with pagination. For chain coverage, edge cases, or response fields, use browse_docs with path 'api/developer/assets/list-assets.md'.
• get_realtime_crosschain_assets — Get cross-chain assets by ID, slug, or chain and token address. For chain coverage, edge cases, or response fields, use browse_docs with path 'api/developer/assets/get-assets.md'.

Answers a handshake

not on the last check · 2026-10-15

Answers a call

not asked on the last check

Live probe verifies the interface responds — not what the tools do, their safety, or their output quality. The call is one read-only tool named by this server's own list (list_/get_/search_, no mutating verb anywhere in the name, no required arguments and none that look like a key, an address or an amount), asked once a day with no arguments. Not asked is not a failure.

Where Allium MCP fits

Related concepts: What Are MCPs? The Standard Way to Give Agents Tools, What Are Onchain Agents?