Three tools with basic schemas and descriptions, but significant gaps in parameter documentation and output schema clarity. All tools are READ_ONLY and well-named with action verbs (get*). Descriptions are present but generic (10-50 chars). Input schemas use Zod with type definitions, but parameter descriptions are minimal or missing context. No output schema documentation. Error handling returns JSON but lacks recovery guidance. Tool composition is sound (each does one thing), but parameter naming could be more explicit (e.g., 'tokenAddress' vs 'token_address').
Get ALL token balances for a wallet address on Berachain
Get information about an ERC20 token on Berachain
Get total supply for an ERC20 token on Berachain
Output schemas not documented. Tools return JSON strings but LLMs cannot infer field structure, types, or required fields. Downstream tool chaining is impossible without trial-and-error.
Parameter descriptions are too brief or missing context. 'Wallet address to check token balances for' (getBalance.address) lacks format guidance (0x prefix? length?). 'ERC20 token contract address' (getTokenSupply, getTokenInfo) does not specify format or validation rules.
Tool descriptions are under 50 characters and lack context on when to use each tool. 'Get total supply for an ERC20 token on Berachain' does not explain when to call this vs getTokenInfo or how the result is used.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 44 | 2026-07-28+ | v2 |
Error responses return JSON with 'status: error' and message text, but do not guide recovery. E.g., 'Failed to get token balance' does not suggest retry, lookup, or validation steps.
Parameter naming uses camelCase (tokenAddress, includeZeroBalances) instead of snake_case (token_address, include_zero_balances). Inconsistent with MCP conventions and harder for LLMs to parse.