MCP server that gives AI agents full Solana trading capabilities — snipe, mirror, analyze, execute
Critical gaps across all dimensions. Tool descriptions are present but generic (avg ~40 chars, well below 194-char baseline). Input schemas exist but lack parameter descriptions entirely, every parameter is undescribed, violating the 100% A+ baseline. No output schemas documented. No error handling guidance. Irreversible operations (buy, sell, close_position) lack confirmation patterns or dry-run support. No tool annotations (readOnlyHint/destructiveHint). Naming is verb-prefixed but descriptions fail to explain WHEN to use each tool or what it returns. Security concern: wallet/private key handling visible in code but no mention of secret injection pattern in tool definitions.
Execute a buy transaction for a token on Pump.fun
Close an open position by selling all tokens
Get SOL balance for a given address
Get bonding curve state for a Pump.fun token
Get summary of all open positions and P&L
Get current position details for a held token
Get current SOL price in USD
Get token metadata and supply information
All 13 tools lack parameter descriptions. Input schemas define types (string, number, boolean) but provide zero guidance on valid ranges, formats, or constraints. E.g., solana_buy accepts 'slippageBps' with no description of valid range (0-10000?), units, or consequences of invalid values.
No output schemas documented for any tool. LLMs cannot plan downstream calls or extract required fields. E.g., solana_get_balance returns a balance but schema does not specify field names, types, or whether it includes metadata like last_update_time.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 37 | 2026-07-28+ | v2 |
Get historical record of all trades executed
Get server health status including RPC connectivity and trading state
Scan for new tokens on Pump.fun with real-time WebSocket monitoring
Execute a sell transaction for a token on Pump.fun
Enable or disable automatic trading based on signals
Irreversible operations (solana_buy, solana_sell, solana_close_position) lack confirmation or dry-run patterns. An agent can execute a $10k trade with no confirmation step, risking catastrophic errors.
No tool annotations (readOnlyHint, destructiveHint, idempotentHint). Agents cannot distinguish safe read-only tools from destructive ones without parsing descriptions. solana_buy and solana_sell should be marked destructiveHint=true.
Tool descriptions are generic and under 50 chars (baseline 194 chars). E.g., 'Get current SOL price in USD' does not explain when to call this vs other price sources, what precision is returned, or whether the price is real-time or cached.