MCP server for SaucerSwap DEX integration on Hedera, enabling token swaps (HBAR↔Token, Token↔Token) and pool data access
Three tools with complete input schemas and descriptions, but quality is uneven. All tools have regex-validated Hedera ID parameters and numeric defaults. Descriptions are adequate (40-60 chars) but lack actionable context about when to use each tool or what happens on failure. No output schema documented. No error handling guidance. Tools follow verb_noun naming but lack distinction in descriptions, LLMs may conflate swap operations. Missing parameter descriptions for some fields (e.g., what 'slippageTolerancePercent' means in practice, why 0.5% default). No mention of prerequisites (e.g., user must have token balance, account setup). Schemas are well-formed with Zod validation, but responses are raw JSON without field documentation.
Swaps HBAR for a specified HTS token
Swaps a specified HTS token for HBAR
Swaps one HTS token for another HTS token
No output schema documented. LLMs cannot predict what fields SwapResult contains or plan downstream operations. Responses are raw JSON strings without field descriptions.
Descriptions lack actionable context. 'Swaps HBAR for a specified HTS token' does not explain when to use this vs swapTokenForToken, what prerequisites exist (account setup, token association), or what happens on failure. Descriptions are 40-60 chars; baseline is 194 chars for A+ tools.
Parameter descriptions are generic and lack constraints. 'slippageTolerancePercent' description does not explain what happens if slippage exceeds tolerance (transaction reverts?), or why 0.5% is the default. 'deadlineSeconds' lacks context about what happens if deadline passes.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 48 | <=2025-11-25 | v2 |
No error handling guidance. Tool implementations throw errors but do not provide recovery hints. LLMs cannot distinguish retryable errors (network timeout) from user-fixable errors (insufficient balance) from fatal errors (invalid token ID).
Tool names lack distinction. All three tools start with 'saucerswap-swap' and descriptions are similar. LLMs may conflate them or pick the wrong one. Consider prefixing with direction: 'saucerswap-swap_hbar_to_token', 'saucerswap-swap_token_to_hbar', 'saucerswap-swap_token_to_token' for clarity.