Three tools with complete schemas and descriptions, but significant gaps in parameter documentation and output schema clarity. ask_compliance has good structure with enum constraints on 'mode', but lacks descriptions for 'site' parameter depth and output format. list_frameworks and get_framework are minimal, no parameter descriptions beyond the bare schema. No error handling guidance, no output schemas documented, and no tool annotations (readOnlyHint). Descriptions are adequate (50-100 chars) but lack WHEN/WHY context. Naming is clear and verb-forward. Composition is sound, each tool has one responsibility.

• Document output schemas for all three tools. For ask_compliance, specify the structure of the response (e.g. {citations: [{source, text}], summary: string}). For list_frameworks, document the tier grouping structure. For get_framework, document the manifest metadata fields.
• Add description to 'site' parameter: 'Comma-separated jurisdiction codes to scope the search (e.g. eu,us-co,uk). If omitted, searches all jurisdictions. Valid codes: eu, us-co, us-ca, uk, ...'
• Add tool annotations to inputSchema: set readOnlyHint=true and idempotentHint=true for all three tools so agents know they are safe to retry without side effects.
• Enhance list_frameworks and get_framework descriptions with discovery context: 'Call list_frameworks first to see available frameworks and their doc_ids, then use get_framework to fetch metadata for a specific framework.'
• Add error handling guidance to ask_compliance description: 'If the query returns no results, try broadening the jurisdiction scope or rephrasing the question. If the corpus is unavailable, call list_frameworks to verify connectivity.'