Panopticon is a security analysis tool for MCP servers, not a production MCP server itself. The 12 tools are test fixtures embedded in the codebase (clean_server, gpt_review_eval, vulnerable_server) used to evaluate MCP behavior. All tools have minimal schemas (only 'path' or 'expression' or generic 'arguments'/'client' objects) with NO descriptions visible in the provided source. Tool names are generic and do not follow verb_noun patterns (e.g., 'justified_reader', 'safe_calculator', 'validated_lookup'). Schemas lack parameter descriptions, type constraints, and output documentation. This is intentional, the fixtures are designed to be analyzed, not used as production tools. Scoring reflects the actual quality of the embedded tool definitions, which are deliberately sparse for testing purposes.

• Add a 1-2 sentence description to every tool explaining what it does, when to use it, and what it returns. E.g., 'justified_reader: Reads file contents from a specified path. Use to retrieve configuration files or documentation. Returns file text as a string.'
• Add descriptions to all input parameters. For 'path' parameters, specify format (absolute/relative), allowed directories, and validation rules. For 'expression' parameters, document the expression language and constraints.
• Replace generic 'arguments' and 'client' object parameters with explicit, typed schemas. Define nested properties, their types, and descriptions. E.g., 'arguments' should expand to {'type': 'object', 'properties': {...}, 'required': [...]}.
• Rename tools to follow verb_noun convention. E.g., 'justified_reader' → 'read_file', 'safe_calculator' → 'evaluate_expression', 'validated_lookup' → 'lookup_value'.
• Document output schemas for each tool. Specify return type (string, object, array), field names, and types. E.g., 'read_file returns {"content": string, "size": integer, "encoding": string}'.