Scoring was not performed

• Rename 'find-tools' to 'recommend_tools' or 'search_available_tools' to follow verb_noun convention and improve LLM intent inference.
• Document the complete output schema: Is it {tools: [{name, description, params}]}? Include field types and purposes.
• Add enum or pattern constraint to the 'prompt' parameter. Consider suggesting prompt structure: 'Describe what you want to accomplish, e.g., "I need to manage container networking" or "Find tools for image scanning."'
• Expand the tool description to explain when to call it: 'Use this tool when the user asks for help finding relevant tools, or when you want to discover available capabilities before making a plan.'
• Document output field requirements: if downstream tools use tool IDs returned here, ensure those IDs are included and named consistently.
• Add error guidance: 'If no tools match, suggest the user refine their request or list tool categories available.'
• Consider whether this tool should be a root-level discovery tool or whether the server should expose domain-specific Docker tools (e.g., create_container, manage_network, scan_image). A meta-tool alone does not provide operational value.