The server exposes a single tool 'onchain_agent_*' with a generic, dynamic naming pattern. No explicit tool registration with schemas is visible in the source code. The description is vague ('Generic onchain agent tool prefix') and does not explain what the tool does, when to use it, or what it returns. No input schema, parameter descriptions, or output schema are documented. The tool name violates verb_noun conventions and uses a wildcard pattern that obscures actual functionality. This is a shell script wrapper around an HTTP endpoint; tool definitions are inferred rather than explicitly registered.

• Register explicit tools with verb_noun names (e.g., 'query_onchain_agent', 'execute_onchain_agent') instead of a dynamic prefix pattern. Each tool should have a fixed name, description, and schema.
• Document input schema for each tool with parameter types, descriptions, and constraints. Example: { 'type': 'object', 'properties': { 'query': { 'type': 'string', 'description': 'The onchain query to execute' } }, 'required': ['query'] }
• Write tool descriptions following the pattern: WHAT it does, WHEN to use it, WHAT it returns. Example: 'Execute an onchain query against the Sato Hub blockchain. Returns structured results with sato_url (canonical page) and verify_url (verification report). Use this to fetch onchain data and verify authenticity.'
• Document the output schema explicitly, including all fields returned (sato_url, verify_url, data structure). Specify pagination if results can be large.
• Add parameter descriptions for all inputs. If the tool accepts a 'query' parameter, explain its format, length limits, and valid values. If it accepts filters or options, document each one.