Server has 4 tools with complete Zod schemas and basic descriptions in Japanese. However, descriptions are minimal (10-30 chars), lack actionable context for LLM selection, and provide no guidance on when to use each tool vs. alternatives. Parameters are typed but lack constraints (e.g., depth 1-3 is mentioned in description but not enforced in schema). No output schema documentation. Error handling returns raw error messages without recovery guidance. Tool names are clear (get_*, search_*, filter_*) but descriptions do not explain distinctions between search_contents and filter_contents, or when to use each.
フィルター検索用のツール
特定コンテンツ取得のツール
コンテンツ一覧取得のツール
コンテンツ検索のツール
Descriptions are too short (10-30 chars) and lack LLM-actionable context. 'コンテンツ一覧取得のツール' (content list retrieval tool) does not explain when to use get_contents vs. search_contents vs. filter_contents, or what each returns.
No output schema documented. LLMs cannot plan downstream calls or extract fields without knowing the response structure (e.g., does get_contents return {contents: [], totalCount: 0} or {items: []}?).
Parameter constraints mentioned in descriptions (e.g., 'depth (1-3)', 'limit max 100') are not enforced in Zod schemas. LLMs cannot parse natural-language constraints reliably; use z.number().min(1).max(3) instead.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 47 | 2026-07-28+ | v2 |
Error handling returns raw error messages ('エラー: ...') without recovery guidance. Per pattern:recovery-guide, errors should suggest next steps: 'Invalid endpoint. Available endpoints: blog, news, products. Try again with a valid endpoint.'
search_contents and filter_contents have overlapping functionality (both filter/search content) but descriptions do not clarify the distinction. LLMs will struggle to choose between them.