Bitcompare MCP Server
OfficialServer Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| BITCOMPARE_API_KEY | Yes | Your Bitcompare API key, starts with ck_live_. Free keys unlock get_rates, get_rate_by_symbol and list_providers; other tools require a paid plan (Starter or above). Get a key at https://auth.bitcompare.net/pro/signup. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| get_ratesA | List current rates across providers, optionally filtered by symbol, category, or provider. For "what is the best yield on BTC right now?" style questions, ALWAYS pass category="lending"/"borrowing"/"staking" — an unfiltered call mostly returns spot prices (category="price"), not yields. |
| get_rate_by_symbolA | Fetch all provider rates for a specific coin symbol. Optionally filter by category — pass category="lending"/"borrowing"/"staking" for yield questions, since unfiltered results mix in spot prices (category="price"). |
| get_rate_historyA | Historical rate timeseries for a symbol. Recent history (~30 days) is per-tick; older history is daily granularity. Depth is clamped to your plan (up to 365 days on Pro) and bounded by how far back data has been recorded. (Current plan: up to 7 days of history.) |
| list_providersA | All rate providers, optionally filtered by product category. |
| get_coinA | Full metadata for a coin: description, links, categories, market data, developer stats. |
| list_coinsA | Paginated list of coins with optional name/symbol search. |
| top_coinsB | Top N coins ordered by market capitalisation. |
| similar_coinsA | Coins related to the given coin by category/sector similarity. |
| coin_historyA | Historical price timeseries for a coin. Requests use the nearest 1, 7, 30, 90 or 365-day window (default 7), bounded by available data. Requires a paid plan (Starter or above) over MCP; this endpoint does not apply the rate-history per-plan day clamp. |
| get_priceA | Current aggregated price for one or more symbols, computed from multiple exchange feeds. |
| get_price_by_symbolA | Aggregated exchange price for one symbol, plus the raw per-provider prices that fed the aggregate. Use get_price for multiple symbols in one call. |
| market_summaryA | One-call summary of total market cap, 24h volume, BTC/ETH dominance, and recent trend. |
| fear_greed_indexA | Current Fear & Greed index value and historical trend. |
| top_moversA | Biggest 24h gainers and losers within a rank segment. |
| stablecoin_indexA | Ranked stablecoin leaderboard with stability scores, peg deviation, and market cap. |
| stablecoin_peg_stabilityB | Peg deviation history and stability stats for a stablecoin. |
| resolve_symbolA | Map an exchange-specific or ambiguous symbol to bitcompare's canonical coin id. Useful when an exchange uses a non-standard ticker. |
| resolve_symbols_batchA | Resolve up to 100 symbols in a single request. Requires a plan with bulk endpoints enabled. Your current plan does not include bulk endpoints — calls will return 403. |
| stablecoin_earn_benchmarkA | The Bitcompare Earn Index for stablecoins: median USDC/USDT lending APY across servable providers, with CeFi/DeFi splits, TVL-weighted variant, deltas, and an immutable daily history. Use for "what does stablecoin yield pay right now / how has it trended?" — this is the citable market reference, not a single provider quote. |
| get_rate_changesA | The immutable rate-change log: every material yield move with full provenance (previous rate, raw value, multiplier, collection method, confidence, base/bonus split). Time-ascending pages via cursor. Unfiltered windows are capped at 31 days — filter by provider or symbol to walk deep history. Use for "what changed since X?" and audit trails. |
| list_earn_productsA | The canonical earn product graph: products (provider × asset × category) with their term variants — lockups, tiers, minimum deposits, payout schedules, withdrawal notice, TVL, promo flags — each carrying collection method, confidence (0-100 with bands), verification timestamps, and geo eligibility resolved from variant overrides or provider defaults. Use for "what are the actual terms behind this rate?" |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 21 tools
Most tools target distinct resources (prices vs. yields vs. coin metadata vs. stablecoin analytics), and descriptions actively steer usage, e.g. telling agents to pass category="lending" and to prefer get_price over get_price_by_symbol for multiple symbols. The main soft spots are get_rates vs. get_rate_by_symbol and get_price vs. get_price_by_symbol, which overlap substantially and are only separated by prose hints.
Mostly snake_case with a predictable verb_noun style for retrieval (get_*, list_*, resolve_*), but the set mixes verbs with bare nouns (top_coins, market_summary, fear_greed_index, stablecoin_index, similar_coins). Analogous operations are named inconsistently, e.g. get_rate_history vs. coin_history.
21 tools is on the heavy side for a data API and covers several domains (coins, prices, yields, providers, stablecoins), so some surface likely goes unused in a single session. Each tool does map to a real endpoint, so it is not padding, but it sits at the borderline of the acceptable range.
Market data, rates/yields, provider listings, earn-product terms, rate-change audit logs, and stablecoin analytics are all covered, so core workflows have no dead ends. Gaps are minor: single-product lookup by id and bulk symbol resolution are missing or plan-gated — resolve_symbols_batch is exposed but explicitly returns 403 on the current plan.