Awesome MCP Server providing tools for cryptocurrency liquidation analysis, HyperLiquid whale tracking, Solana token information, and Vietnam stock market data
This server demonstrates moderate definition quality with consistent patterns across 18 tools. Strengths: all tools have descriptions (8-165 characters), organized into coherent domains (liquidation, whale tracking, token analysis, Vietnam stock data), and input schemas are present with type definitions and descriptions for most parameters. Weaknesses: descriptions are brief (average ~95 chars), often lack actionable context about when to use the tool vs. alternatives, no documented output schemas, no error recovery guidance, and parameter descriptions could be more prescriptive about constraints and formats. For example, 'get_liquidation_heatmap' says 'Get the liquidation heatmap data' but doesn't explain what a liquidation heatmap is, when an LLM should call it instead of 'get_liquidation_map', or what the response structure looks like. Parameters like 'symbol' have examples ('BTCUSDT, ETHUSDT') in descriptions rather than formal constraints. No tools declare output structure, pagination behavior, or error cases. Token count efficiency is reasonable but descriptions could be more LLM-optimized.
Get the current position of a specific address on the HyperLiquid platform.
Get the transaction/position history of a specific address on the HyperLiquid platform.
Get the total profit/loss (PnL) of a specific address on the HyperLiquid platform.
Get company profile information of a Vietnam stock ticker.
Get financial statement data of a Vietnam stock ticker.
Get fundamental analysis data of a Vietnam stock ticker.
Output schemas not documented. Tools return JSON via `{ content: [{ type: 'text', text: JSON.stringify(data) }] }` but there is no schema specification describing what fields the actual data object contains, their types, or how pagination/metadata is structured. LLMs cannot plan downstream operations (e.g., which field holds the address to pass to another tool) without knowing the response structure.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | D | 52 | 2026-07-28+ | v2 |
| 2026-03-09 | F | 47 | - | v1 |
Get historical price data of a Vietnam stock ticker.
Get the latest whale actions on the HyperLiquid platform. Provide information about the large transactions and notable activities.
Get the liquidation heatmap data of a cryptocurrency pair. Provide information about the price levels with a high concentration of liquidation orders, represented as a heatmap.
Get the liquidation map data of a cryptocurrency pair. Provide detailed information about the liquidation positions by price level.
Get major shareholders/holders information of a Vietnam stock ticker.
Get the list of official posts from Vietnam stock market on Fireant platform.
Get the bubble map token's holders on the Solana blockchain. We can identify the token's holders and their transaction history and check the healthy distribution of the token.
Get the token info of a specific address and check rug pull risk on the Solana blockchain.
Get the price of a specific token on the Solana blockchain.
Get the list of user posts from Vietnam stock market on Fireant platform.
Get the statistics of the profit/loss (PnL) of the whales on the HyperLiquid platform.
Get the list of whale positions on the HyperLiquid platform.
Descriptions lack actionable context. Many descriptions are under 70 characters and do not explain WHEN to use this tool vs. similar ones (e.g., what distinguishes 'get_liquidation_heatmap' from 'get_liquidation_map'?), what the output contains, or what prerequisites exist. Example: 'Get the total profit/loss (PnL) of a specific address on the HyperLiquid platform.' does not explain if this is cumulative, what time period it covers, or whether it includes unrealized gains.
Parameter descriptions include example values ('e.g: BTCUSDT, ETHUSDT') instead of formal constraints. LLMs often reuse example values literally rather than adapting to context. 'symbol' should constrain format (uppercase, format /^[A-Z0-9]+USDT?$/) rather than just giving examples.
No error handling or recovery guidance. Tools do not document what error conditions may occur (invalid symbol, rate limit, service unavailable), whether errors are retryable, or what the LLM should do next. A failed call to 'get_solana_token_info' with an invalid address should guide the LLM to retry with a valid Solana address format, but there is no such guidance.
Pagination parameters ('page', 'size', 'offset', 'limit') are optional but have no documented defaults or constraints on ranges. What is the maximum page size? Does offset=0 mean the first record or one-indexed? Are there bounds to prevent requesting 1 million records?
No tool-chaining hints. When 'get_whale_positions' returns positions, does it include an 'address' field that can be passed to 'get_address_total_pnl'? Without documented response fields, LLMs cannot infer the chain of operations needed to compose tools.