Isthmus provides 30 tools with consistent naming (verb_noun pattern) and descriptions. All tools have input schemas with type definitions. However, descriptions are often terse (10-50 chars), parameter descriptions are sparse or missing, and output schemas are undocumented. Error handling guidance is absent. The server follows basic structure but lacks LLM-optimization depth expected of production tools.

• Expand tool descriptions to 100-200 chars. Include WHAT the tool does, WHEN to use it, and what it returns. E.g., 'health' → 'Check network adapter health across MeshCore, Meshtastic, Reticulum, Nostr, and ACP agent. Returns status (online/offline) and last-seen timestamp for each. Call this first to verify connectivity before sending messages.'
• Document output schemas for all tools. Specify field names, types, and whether results are paginated. E.g., list_groups should return {groups: [{id, name, members_count, created_at}], total_count, next_cursor}.
• Add parameter descriptions for all enum-like fields. E.g., 'network' in attach_member → 'Network type: meshcore | reticulum | nostr | agent. Determines which identity format is expected (pubkey, npub, dest hash, or agent name).'
• Add recovery hints to error-prone tools. E.g., get_group → 'If group not found by name, call list_groups() to see available groups and their UUIDs.'
• Implement confirmation or dry-run for destructive tools. E.g., revoke_group should accept a 'confirm=true' parameter and return a preview of what will be deleted before requiring explicit confirmation.

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

• Relies on Stateful initialize / Mcp-Session-Id (removed; protocol is stateless) - make each request self-contained