Server has 7 tools with consistent naming (verb_noun pattern: fireplace_list_styles, fireplace_get_html, etc.) and reasonable descriptions (100-200 chars each). All tools have input schemas with types and enums. However, parameter descriptions are minimal (10-30 chars), output schemas are not documented, and error handling is absent. No tool annotations (readOnlyHint, destructiveHint). Descriptions lack WHEN-to-use context and dependency hints. Parameter constraints are present (enums for response_format) but descriptions don't explain format expectations or valid slug values. No pagination despite list_styles returning 34 items. Baseline: 194 chars avg description, 72 chars avg param annotation, this server averages ~120 chars descriptions and ~20 chars param annotations.

• Expand parameter descriptions to 50-100 chars. E.g. 'slug: Fireplace identifier from fireplace_list_styles. Must be lowercase alphanumeric with hyphens (e.g. classic-hearth, wood-stove).'
• Document output schemas in tool descriptions. E.g. 'Returns array of {slug, name, mood, palette, tags[], audio: bool}' for fireplace_list_styles.
• Add readOnlyHint: true to all 7 tool schemas to signal safety to agents.
• Add limit parameter to fireplace_list_styles (default 20, max 34) and return {fireplaces: [...], total: 34, limit: 20}.
• Add error recovery hints: 'If slug is invalid, call fireplace_list_styles to discover valid options.' Include this in tool descriptions.