Astro integration that exposes your site content via WebMCP for AI agents
Three tools with clear verb-noun naming (search_content, list_sections, go_to) and reasonable descriptions (50-80 chars). All have input schemas with types. However, parameter descriptions are minimal or absent, output schemas are undocumented, and error handling guidance is missing. search_content lacks description for 'collection' and 'limit' parameters. No documentation of what fields are returned or how pagination works. Follows basic structure but lacks LLM-optimization depth expected of production tools.
Navigate to a specific page on this site by its slug.
List all content sections (collections) available on this site with item counts.
Search articles and pages on this site by keyword. Returns title, URL, and description of matching results.
Output schemas undocumented. No specification of what fields search_content, list_sections, or go_to return. LLMs cannot plan downstream tool calls or extract required data without knowing response structure.
Parameter descriptions missing or incomplete. 'collection' and 'limit' in search_content lack descriptions. 'slug' in go_to has minimal context. LLMs cannot infer parameter meaning from names alone.
No pagination guidance. search_content accepts 'limit' but no documentation of default, max, or whether results are paginated. Large result sets could exhaust context window.
Error handling absent. No guidance on what errors can occur (e.g., invalid slug, no results) or how LLM should recover. Raw errors provide no actionable next steps.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 60 | 2025-06-18+ | v2 |
go_to tool marked REVERSIBLE but no confirmation or dry-run pattern documented. Navigation may have side effects (analytics, state changes) that should be confirmed before execution.