Deployable WebMCP and MCP sidecar for public Wagtail content. Exposes Wagtail CMS pages to AI agents over MCP, llms.txt and WebMCP.
Server exposes 4 read-only tools via HTTP with basic schemas and descriptions. All tools have names starting with action verbs (get_, list_, search_) and descriptions present. However, parameter descriptions are minimal or absent, output schemas are not documented, and error handling lacks recovery guidance. The implementation is functional but falls short of production-grade quality. No tool annotations (readOnlyHint, etc.) despite all being read-only operations.
Retrieve the full content of a specific page by URL
Get the list of all available pages on the site
List all available pages with their metadata
Search for pages by query string
Output schemas not documented. LLMs cannot predict response structure, forcing them to guess at field names and types for downstream reasoning.
Parameter descriptions missing or minimal. 'uri' in get_page_content lacks format guidance (is it a full URL or path?). 'limit' in search_site has no range constraints (1-100? 1-1000?).
No tool annotations despite all being read-only. readOnlyHint should be applied to all 4 tools to signal safety to agents.
Error handling lacks recovery guidance. No indication of what to do if a page is not found, Wagtail API is down, or search returns no results.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 64 | 2026-07-28+ | v2 |
get_sitemap and list_content have identical purpose but different names. LLMs will waste reasoning cycles deciding between them. Consolidate to one canonical tool.