Three tools with basic schemas and minimal descriptions. All tools have input schemas (Zod-validated), but descriptions are generic and lack context for LLM selection. Parameters lack descriptions entirely, 'projectName' and 'path' have no guidance on format, constraints, or when to use them. No output schema documentation. Error handling returns JSON but provides no recovery guidance. Tool names follow verb_noun convention (list_*, read_*) but descriptions don't explain WHEN to call each tool or what distinguishes them. No pagination, no field documentation, no chaining IDs for downstream calls.

• Add 50-150 character descriptions to each tool explaining WHEN to call it and what it returns. E.g., 'Lists all Unity projects registered in Unity Hub. Call this first to discover available projects before listing files.'
• Add descriptions to all parameters. E.g., 'projectName: The display name of the Unity project (e.g., "MyGame"). Must match exactly as shown in list_projects output.'
• Document output schemas. E.g., list_projects returns [{title: string, path: string, version: string, ...}]. read_file returns {content: string, encoding: string}.
• Enhance error messages with recovery guidance. E.g., 'Project "MyGame" not found. Available projects: [list]. Call list_projects() to see all projects.'
• Add pagination to list_files_in_project: accept limit (default 50) and offset parameters, return {files: [...], total: number, hasMore: boolean}.