Brain Kit MCP has 4 tools with clear action verbs and reasonable descriptions (avg 120 chars). All tools have input schemas with Zod validation. However, parameter descriptions are sparse or missing: 'limit', 'threshold', 'days' lack descriptions entirely. Output schemas are not documented, LLMs cannot predict response structure. Error handling returns text but lacks recovery guidance. Tool composition is sound (search, list, stats, capture), but missing pagination metadata (total_count, next_cursor) on list results.

• Add descriptions to all parameters: 'limit' (max results, default 10), 'threshold' (similarity score 0 - 1, default 0.5), 'days' (lookback window in days), 'type' (enum: observation|task|idea|reference|person_note), 'topic' (tag string), 'person' (name string).
• Document output schemas for each tool. E.g., search_thoughts returns array of {content, metadata, similarity, created_at}; list_thoughts returns {total, items[], has_more}.
• Add pagination to list_thoughts: return total_count and next_cursor; accept cursor parameter for continuation.
• Enhance error responses with recovery hints: 'Search failed: embedding service unavailable. Retry in 30s or try list_thoughts() instead.'
• Document capture_thought's auto-extraction: 'Returns {id, created_at, metadata: {type, topics, people, action_items, dates_mentioned}}' so LLM knows what to expect.