Three tools with complete input schemas and descriptions. Naming follows verb_noun pattern (beck_search, beck_get_document, beck_get_toc). Descriptions are adequate (100-150 chars) but lack actionable guidance on when to use each tool or dependencies between them. Parameters are typed with enums where appropriate (documentType, legalArea, sort), but descriptions are minimal (10-30 chars). Output schemas are not documented, LLMs cannot predict response structure. No error handling guidance visible. No tool annotations (readOnlyHint, etc.). Composition is sound: three single-purpose tools with clear separation. Missing: pagination guidance, response field documentation, error recovery hints.

• Document output schemas for each tool. Example for beck_search: 'Returns {results: [{title, documentType, vpath, snippet}], totalCount, page}'. This enables LLMs to extract vpath for downstream calls.
• Expand parameter descriptions to 50-100 chars with enum meanings. E.g., 'Filter by document type: ges=laws, buch=commentaries, aufs=articles, form=forms, sonst=other, lex=lexicon, verwan=related'.
• Add tool-chaining hints in descriptions. E.g., beck_search: 'Returns vpath values for use with beck_get_document.' beck_get_document: 'Requires vpath from beck_search results.'
• Add error recovery guidance. E.g., 'If no results, try a shorter query or remove legalArea filter. If document not found, verify vpath from search results.'
• Add readOnlyHint annotations to all three tools in registration code to signal safety to agents.