Both tools have basic descriptions but lack critical schema documentation. Input parameters are visible in code but not formally declared with JSON Schema types/constraints. Descriptions are present but generic (under 200 chars). No output schemas documented. No error handling guidance. Parameter descriptions exist but lack format/constraint details. Tool names follow verb_noun pattern (positive), but overall quality is well below production baseline.

• Add formal JSON Schema to both tools with type declarations (string, object, array), required fields, and constraints. Example: url must match regex '^https://(bytedance\.)?feishu\.(cn|com)/docs/docc/[a-zA-Z0-9_-]+$'.
• Document output schemas explicitly. Define return object structure: {success: boolean, title?: string, author?: string, content?: string, images?: string[], word_count?: number, error?: string}.
• Expand tool descriptions to 150-250 chars. Explain WHEN to use each: 'read_feishu_doc fetches full content + images; use when you need complete document text. get_doc_info retrieves metadata only; use for quick checks without downloading.'
• Add error classification and recovery hints. Example: 'Invalid URL format (must be feishu.cn/docs/docc/...). Try get_doc_info() first to verify URL is accessible.'
• Formalize cookies_str parameter: describe format as 'semicolon-separated key=value pairs (e.g., "session=abc123; token=xyz789")' and note that it is optional and only needed for authenticated documents.

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

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