MCP server for WordPress sites running Bricks Builder, enabling AI assistants to query and modify site content, pages, elements, templates, media, and theme settings.
LC Bricks MCP exposes 2 tools with basic schemas and descriptions. Both tools have input schemas with type definitions and enum constraints, which is positive. However, descriptions are generic and lack LLM-optimized guidance on WHEN to use each tool or WHAT to do with results. The 'wordpress' tool conflates 4 distinct actions (get_posts, get_post, get_users, get_plugins) into a single tool with an action enum, this violates single-responsibility and forces the LLM to reason about which action to invoke. Parameter descriptions exist but are minimal (avg ~40 chars vs baseline 72 chars). No output schemas are documented. Error handling is absent from visible code. The parity checker (bin/check-tool-parity.php) is a strong engineering signal, but does not compensate for weak LLM-facing definitions.
Get WordPress site information. Actions: - info: Basic site info (default) - diagnose: Run connection diagnostics (REST API, App Passwords, security plugins, hosting provider, MCP endpoint)
Query WordPress data. Actions: - get_posts: List published posts (optional: post_type, posts_per_page, orderby, order) - get_post: Get single post/page by ID (requires: id) - get_users: List users (no required params) - get_plugins: List plugins (no required params)
Tool 'wordpress' violates single-responsibility: combines get_posts, get_post, get_users, get_plugins into one tool with action enum. LLMs must reason about which action to invoke; should be 4 separate tools (get_posts, get_post, get_users, get_plugins).
No output schemas documented. LLMs cannot plan downstream calls or extract required fields. E.g., does get_posts return post_id, post_title, post_content? Does get_users return user_id, user_email, user_role?
Descriptions lack LLM-optimized guidance. 'Get WordPress site information' (36 chars) and 'Query WordPress data' (20 chars) are too generic. Should state WHEN to use, WHAT it returns, and any prerequisites. Baseline: 194 chars avg.
Parameter descriptions are minimal. E.g., 'Action to perform' (19 chars) lacks context. Should explain each action's purpose and when to use it. Baseline: 72 chars avg per param.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 50 | 2026-07-28+ | v2 |
No error handling guidance visible. If a post ID is invalid or a user role filter fails, what does the tool return? How should the LLM recover? Missing recovery guides per pattern:recovery-guide.