MCP server for interacting with Marinade Finance protocol on Solana, providing tools for documentation search, protocol state queries, and on-chain operations like staking and balance checks.
Three tools with basic descriptions and minimal schemas. search_documentation and get_marinade_state have reasonable descriptions (100-150 chars), but get_msol_balance is under-described (80 chars). No output schemas documented. Error handling returns JSON but lacks recovery guidance. Tool names follow verb_noun pattern (good), but parameter naming is inconsistent (walletAddress uses camelCase; no type hints in descriptions).
Retrieve the current state of the Marinade Finance protocol, including information about staked assets, mint address, price, rewards, and other relevant data.
Check the mSOL token balance of the environment wallet or any specified Solana wallet address.
Search across the documentation to find relevant information, code examples, API references, and guides. Use this tool when you need to answer questions about Marinade Finance Docs, find specific documentation, understand how features work, or locate implementation details. The search returns contextual content with titles and direct links to the documentation pages.
Input parameter descriptions missing. 'query' in search_documentation and 'walletAddress' in get_msol_balance lack descriptions explaining format, constraints, or examples.
Output schemas not documented. Tools return JSON text blobs without declaring field structure, types, or required fields. LLMs cannot plan downstream calls or extract data reliably.
Error handling lacks recovery guidance. Errors return JSON with 'error' and 'reason' fields but do not tell the LLM what to do next (retry, ask user, call another tool).
Inferred effective spec: 2026-07-28+.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 46 | 2026-07-28+ | v2 |
get_msol_balance description is 80 chars and generic ('Check the mSOL token balance...'). This one lacks context for tool selection.
get_marinade_state has empty inputSchema ({}). While this is valid for a no-argument tool, the tool definition does not declare what fields the response contains, making it impossible for LLMs to reason about the output.