Five tools with complete input schemas and descriptions. Naming follows verb_noun pattern (feishu_reply, feishu_send_markdown, feishu_allowlist, feishu_pair). Descriptions are adequate (80-180 chars) but lack actionable context for LLM selection. Parameters are typed and mostly described, but several lack format/constraint guidance. No output schemas documented. Error handling is minimal, no recovery guidance or categorization. feishu_allowlist has a conditional parameter (open_id required only for add/remove) that is underdocumented. feishu_pair lacks context on how to obtain the pairing code.

• Document output schemas for all five tools. Specify what fields are returned (e.g., success: boolean, message_id: string, error?: string) so LLMs can chain tools and extract results.
• Add conditional parameter validation to feishu_allowlist: mark open_id as required only when action is 'add' or 'remove'. Update the description to state: 'open_id is required for add/remove actions, omit for list.'
• Enhance feishu_pair description: 'Complete pairing via 5-char code sent to your Feishu DM by the bot when an un-allowlisted sender tries to message. Run this tool with that code to authorize them. Codes expire after 10 minutes.'
• Add error recovery guidance to reply/send tools: 'If message_id is invalid, the message may have been deleted. Try searching recent messages in the chat to find a valid message_id.'
• Add format constraints to parameter descriptions: 'message_id must be a Feishu message ID starting with om_ (e.g., om_abc123def456). chat_id must start with oc_ (e.g., oc_xyz789).' Consider adding regex patterns to the schema.