Server has 4 tools with explicit schemas and descriptions, but quality is inconsistent. Tool naming follows verb_noun convention (good). However, descriptions are minimal (10-50 chars), parameter descriptions lack detail, and output schemas are not documented. clickup_create_task has the most complete schema with enums for priority; others are sparse. No error handling guidance, no pagination support despite list operations, and no indication of which tools modify state. Schemas are present but lack depth, many parameters have generic descriptions like 'List ID to get custom fields for' without format/constraint guidance.

• Return structured JSON objects instead of formatted text. clickup_get_lists should return {lists: [{id, name, folder_id, ...}]} so LLM can extract IDs for chaining.
• Add format/constraint descriptions to all parameters: 'list_id (string, 10-20 alphanumeric chars)', 'name (string, 1-255 chars)', 'priority (enum: urgent|high|normal|low)'.
• Implement pagination: add limit (default 20, max 100) and offset/cursor parameters to clickup_get_lists and clickup_get_spaces. Document in tool description.
• Add error handling wrapper that catches httpx exceptions and returns structured errors: {error: 'not_found', message: 'List not found. Try clickup_get_lists() to find valid IDs.', retryable: false}.
• Prefix tool descriptions with action verbs and state modification intent: 'Creates a new task in ClickUp (modifies state)' instead of 'Create a new task/story in ClickUp'.