Wraith exposes 2 tools for resource discovery and reading. Both tools have descriptions in Chinese, but they are minimal (under 100 chars) and lack critical details about when to use them, what they return, and error handling. Input schemas are present but sparse: list_resources has an empty properties object (no parameters), and read_resource has only a uri string parameter with minimal description. No output schemas are documented. No error handling guidance, no parameter validation rules, and no indication of pagination or result limits. The tools appear to be meta-tools for introspecting the MCP server itself rather than domain-specific functionality.

• Expand tool descriptions to 100-200 characters in English. For list_resources: 'Discover all resources exposed by this MCP server. Returns a list of resource URIs, names, MIME types, and descriptions. Call this first to see what data is available before reading specific resources.'
• Document output schemas for both tools. For list_resources, specify: {type: 'object', properties: {resources: {type: 'array', items: {type: 'object', properties: {uri: {type: 'string'}, name: {type: 'string'}, mimeType: {type: 'string'}, description: {type: 'string'}}}}}}. For read_resource, specify: {type: 'object', properties: {uri: {type: 'string'}, contents: {type: 'string'}, mimeType: {type: 'string'}}}.
• Add error handling guidance to read_resource description: 'If the URI is invalid or resource not found, the tool returns an error with available resource URIs. Use list_resources to find valid URIs.'
• Add parameter constraints to read_resource: 'uri must be a valid resource URI returned by list_resources or explicitly provided by the user. Format: scheme://path (e.g., file:///path/to/resource).'
• Translate descriptions to English and ensure they are discoverable by English-speaking LLMs. Chinese descriptions limit tool adoption and LLM understanding.