EcoRelay defines 19 tools with complete input schemas and descriptions. Naming follows verb_noun convention (relay_*) consistently. However, descriptions are brief (avg ~60 chars, well below the 194-char baseline for A+ tools) and lack context on WHEN to use each tool or what happens on failure. Parameter descriptions are present but minimal. Output schemas are not documented, callers must infer structure from code. Error handling returns generic codes ('bad_args', 'unexpected') without recovery guidance. No tool annotations (readOnlyHint, destructiveHint) despite clear risk stratification in the data. Schemas are well-formed JSON Schema with proper types and constraints (maxLength, required arrays), but lack examples or detailed format guidance.

• Add tool annotations to all tools. Use readOnlyHint for relay_peers, relay_inbox, relay_rooms, relay_group_history, relay_group_list, relay_group_info. Use destructiveHint for relay_group_delete. Use idempotentHint for relay_rename (if safe to retry).
• Expand descriptions to 100 - 200 chars. Include WHEN to use (e.g., 'Use relay_broadcast to ask all peers a question; use relay_send for direct peer-to-peer messages'). State whether calls are fire-and-forget or blocking.
• Document output schemas in tool descriptions or a separate schema registry. E.g., 'Returns {ok: boolean, msg_id: string, status: string}' for relay_send. Include field types and when fields are optional.
• Replace generic error codes with actionable messages. E.g., instead of 'bad_args', return 'Invalid recipient: to must be 1 - 64 chars, got 128. Did you mean: [list of peers]?' or 'Message too long: max 4096 chars, got 5000.'
• Add explicit pagination to relay_group_history (add 'limit' param, document max results). Clarify whether relay_peers, relay_rooms, relay_group_list have result caps and how to fetch more.

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

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