Four tools with complete schemas and descriptions, but significant gaps in parameter documentation and error handling. Tool names follow verb_noun convention (reply, react, download_attachment, edit_message). Descriptions are present (avg ~80 chars) but lack actionable context for LLM selection. Parameters have types but many lack descriptions (e.g., 'format' enum in reply/edit_message has no description explaining when to use markdownv2 vs text). No output schemas documented. Error handling is minimal, no recovery guidance or categorization. Composition is reasonable (each tool does one thing), but missing idempotency guarantees and chaining metadata.

• Add descriptions to all enum parameters. For 'format' in reply/edit_message: 'Rendering mode: text (plain) or markdownv2 (bold, italic, code blocks). Default: text.'
• Document output schemas for each tool. E.g., reply should return {message_id: string, chat_id: string, timestamp: number} so downstream tools can reference the sent message.
• Add error handling guidance. E.g., reply description: 'Returns error if chat_id is invalid or user is not allowlisted. Check access.json or call router to pair new users.'
• Clarify when to use reply vs edit_message in descriptions. E.g., reply: 'Send a new message.' edit_message: 'Modify a previously sent message (no notifications).'
• Document idempotency and side effects. E.g., react: 'Idempotent, adding the same emoji twice has no effect. Emoji must be a single Unicode character or emoji sequence.'