Four tools with basic schemas and descriptions, but significant gaps in parameter documentation, output schema clarity, and error handling. Tool names follow verb_noun convention (analyze_, draft_, check_, assess_) but descriptions are generic (avg 80 chars, below 194-char baseline). Parameters lack descriptions for complex objects (parties, key_terms in draft_contract_belgian). No documented output schemas, pagination support, or error recovery guidance. No tool annotations (readOnlyHint/destructiveHint). Schemas present but incomplete, object parameters lack property definitions.

• Add explicit JSON Schema property definitions for object parameters. For draft_contract_belgian, define parties as {type: 'object', properties: {employer: {type: 'string'}, employee: {type: 'string'}, ...}, required: ['employer', 'employee']} and key_terms similarly.
• Document output schemas for all tools. E.g., analyze_contract_clauses should return {type: 'object', properties: {clauses: {type: 'array', items: {type: 'object', properties: {name: string, risk_level: string, ...}}}, summary: string, ...}}.
• Expand parameter descriptions to 50-200 chars explaining WHEN to use each value. E.g., 'contract_type: Type of contract. Use "employment" for employer-employee agreements, "service" for vendor/client work, "nda" for confidentiality agreements. Affects clause extraction and compliance checks.'
• Add error handling with recovery guidance. Catch page-limit violations and return: 'Contract exceeds 20-page limit (Lite edition). Upgrade to Solo (50 pages) or Pro (unlimited), or split the contract into smaller sections and analyze separately.'
• Add tool annotations: @readOnlyHint for analyze_contract_clauses, legal_compliance_check, contract_risk_assessment; @destructiveHint for draft_contract_belgian. This signals to LLMs which tools are safe to retry.

Inferred effective spec: <=2025-11-25.

• Relies on Logging (deprecated) - log to stderr or use OpenTelemetry