Server has 6 tools with clear verb-noun naming (get_, list_, discover_, send_, get_task_status, get_agent_card). Descriptions are present and moderately detailed (100-250 chars), explaining WHAT and WHEN. Input schemas are visible with type declarations. However, parameter descriptions are sparse or missing in several tools (e.g., send_task's 'target' param lacks format/constraint guidance; get_node_info and list_connected_peers have empty input schemas with no description of what they return). Output schemas are not documented, LLMs cannot infer response structure. Error handling is absent from descriptions. No tool annotations (readOnlyHint/destructiveHint). Composition is sound: tools chain logically (discover_agents → send_task → get_task_status). Naming is consistent and action-oriented.

• Document output schemas for all tools. Example: get_node_info should return {peer_id: string, did: string, connection_status: string}. Use JSON Schema in tool definition or add 'Returns:' section to docstring.
• Add parameter descriptions for send_task's 'target' field: 'PeerID (starts with 12D3KooW), skill name (e.g. translate), or HTTP URL (http://...)'. Include format constraints and examples.
• Add tool annotations via FastMCP: mark get_* and list_* tools with readOnlyHint=true; mark send_task with destructiveHint=true. This signals safety to agents.
• Enhance error handling in descriptions. Example for send_task: 'Returns error if target not found (try discover_agents first) or timeout exceeded (retryable; increase timeout param).'
• Add pagination support to list_connected_peers and discover_agents: include limit (default 20, max 100) and offset parameters; return total_count in response.