Server has 4 tools with basic schemas and descriptions, but significant gaps in parameter documentation and error handling. Tool names follow verb_noun convention (launch-token, buy-token, birdeye-trending-tokens, birdeye-top-traders), which is good. However, parameter descriptions are minimal (1-2 words), and critical parameters like 'slippage' lack range constraints. No output schemas documented. Error handling is absent, no recovery guidance or actionable error messages visible in code. The token launcher and buyer tools perform irreversible financial operations (WRITE risk) but lack confirmation patterns or dry-run support. Descriptions are adequate (50-100 chars) but could be more specific about prerequisites and return values.
Get top traders on Solana from BirdEye (showing today's data)
Get trending tokens on Solana from BirdEye
Buy tokens on Solana using the Raydium launchpad
Launch a new meme token on Solana using Raydium launchpad
Destructive tools (launch-token, buy-token) lack confirmation/dry-run pattern. No mechanism to prevent accidental token creation or fund loss.
Parameter descriptions are minimal (1-2 words). 'slippage' has no range constraint; 'amount_sol' has no min/max. LLMs cannot infer valid ranges.
No output schemas documented. LLMs cannot plan downstream calls or extract required fields (e.g., token_address from launch-token response).
Error handling absent. No recovery guidance, actionable error messages, or error classification (retryable vs fatal). Stack traces would confuse LLMs.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 49 | 2026-07-28+ | v2 |
BirdEye tools (trending-tokens, top-traders) have no parameters but descriptions don't explain what data is returned or when to call them.