MCP server for crypto and finance data via CoinGecko
Server has 6 tools with complete input schemas and descriptions. Naming follows verb_noun convention (get_*, compare_*). Descriptions are present and contextual (avg ~120 chars), meeting baseline minimums. However, output schemas are not documented in the visible code, LLMs cannot predict return structure. Parameter descriptions are adequate but lack format constraints (e.g., coinId accepts 'bitcoin' but no enum or pattern shown). Error handling is minimal; no recovery guidance or actionable error messages visible. Tool composition is sound (each does one thing), but missing pagination for list results and no documented limits on returned data.
MUST use this tool whenever the user asks to compare two cryptocurrencies, mentions two coin names together (e.g. 'ETH and SOL', 'Bitcoin vs Ethereum'), or asks which coin is better. Provides side-by-side price, market cap, volume, and 24h performance.
Fetches live market data for a coin, then uses AI (via sampling) to generate a brief market insight. Requires a sampling-capable client.
Get the price history of a cryptocurrency over a number of days, with high/low/open/close summary.
Get the current price, market cap, 24h volume, and 24h price change for a cryptocurrency. Use coin IDs like 'bitcoin', 'ethereum', 'solana', 'cardano'.
Get a global overview of the crypto market: total market cap, BTC dominance, 24h volume, and number of active coins.
Get the top cryptocurrencies ranked by market cap, with current prices and 24h changes.
Output schemas not documented. LLMs cannot predict return structure (fields, types, nesting). Tools return CoinGecko API responses but no schema visible in code.
No pagination or result limits documented. get_top_coins accepts limit 1-25 but no total_count or next_cursor in response. Large result sets risk context window exhaustion.
Error handling lacks recovery guidance. geckoFetch() throws generic errors ('CoinGecko API error: 429') without actionable next steps for LLM (e.g., 'retry after 60s' or 'call search_coins() to validate ID').
Inferred effective spec: 2026-07-28+.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 57 | 2026-07-28+ | v2 |
coinId parameter lacks enum or pattern constraint. Description says 'e.g. bitcoin, ethereum' but no formal validation. LLMs may hallucinate invalid IDs like 'btc' or 'eth' instead of canonical 'bitcoin'/'ethereum'.
get_ai_insight relies on deprecated Sampling pattern (server calls createMessage() to invoke client LLM). Current spec integrates LLM calls directly; Sampling removed ~2027-07-28.