Server has 4 tools with complete input schemas and descriptions. However, descriptions are minimal (10-20 chars baseline), lack actionable context for LLM selection, and omit critical details like prerequisites, return structure, and error guidance. Parameter descriptions exist but are sparse. No output schemas documented. Error handling returns generic text without recovery guidance. Tools lack idempotency hints and confirmation patterns for destructive operations (publish_article is a WRITE operation). Naming is verb-forward but descriptions do not explain WHEN to use each tool or dependencies between them.

• Expand tool descriptions to 80 - 150 characters. Example: 'publish_article: Publish Markdown content to multiple platforms (WeChat, Zhihu, Twitter, etc.). Supports draft mode. Call check_auth first if unsure of platform credentials. Returns task_id for status polling.'
• Document output schemas in tool descriptions or as a separate schema field. E.g., 'Returns {task_id: string, results: [{platform: string, status: string, post_url?: string, error?: string}]}'
• Add tool annotations: mark publish_article with destructiveHint=true to signal irreversible side effects. Mark list_platforms and check_auth with readOnlyHint=true.
• Implement structured error responses with error codes and recovery hints. E.g., 'error_code: AUTH_FAILED, message: Platform authentication required, recovery: Call check_auth(platform) to verify credentials, then retry.'
• Add parameter descriptions explaining platform content-type compatibility. E.g., 'platforms: Target platforms. Note: bilibili_video, douyin, tiktok, kuaishou require video content; others accept articles. Call list_platforms() to see auth status per platform.'

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

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