MCP Server for fetching and exporting Bricks Builder layouts and components from brixies.co
Three tools with strong naming (verb_noun pattern), comprehensive parameter schemas with type constraints and descriptions, and clear output documentation. Tool descriptions are well-written (150-250 chars) and explain WHEN to use each tool. Parameter constraints (enums, minLength, maxLength, patterns) are properly declared. However, output schemas are documented in prose rather than formal JSON Schema, and error handling guidance is minimal. No tool annotations (readOnlyHint/destructiveHint) despite clear risk profiles.
Fetch a layout or component from brixies.co as Bricks Builder JSON, ready to insert into a page via bricks_set_page_elements. Optionally renames CSS class prefixes to avoid collisions with existing classes. The returned bricks_json contains a 'content' array (elements) and 'globalClasses' array.
Search brixies.co layouts and components by keyword. Matches against slug and category (case-insensitive). Results include a preview_url for visual assessment. When snapshots have been synced (via brixies_sync_previews), results also include a snapshot_path for instant local preview — no Playwright screenshot needed. Use the slug with brixies_export to fetch the Bricks JSON.
Fetch the latest sitemap from brixies.co and capture preview screenshots for any newly added layouts or components. Already-cached snapshots are skipped. Removes manifest entries for items that no longer exist in the sitemap. Use this before searching if you want fast, local preview images instead of live Playwright screenshots.
Output schemas documented in prose only; no formal JSON Schema for return types. LLMs cannot parse expected response structure programmatically.
Tool annotations missing. brixies_search and brixies_export are READ_ONLY; brixies_sync_previews is WRITE. Annotations (readOnlyHint, destructiveHint) should be declared in tool metadata for agent planning.
Error handling lacks recovery guidance. No examples of what errors look like or how to recover (e.g., 'Session expired, check BRIXIES_EMAIL and BRIXIES_PASSWORD').
brixies_export 'prefix' parameter has regex pattern but description does not explicitly state the constraint. LLMs may not parse JSON Schema patterns, constraint must be in description text.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 59 | 2025-06-18+ | v2 |
No pagination support in brixies_search. Default limit=20, max=100 is reasonable, but no 'total_count' or 'next_cursor' documented. Large result sets could exhaust context.