This server acts as a proxy to an upstream Habitify MCP server. Only 2 tools are defined: list_tools (READ_ONLY) and call_tool (WRITE). Both have minimal descriptions and the call_tool schema is extremely generic. list_tools lacks any input schema documentation. call_tool's 'arguments' parameter is typed as 'object' with no structure, validation, or guidance on what fields are valid. No output schemas are documented for either tool. Error handling is absent, no guidance on recovery, retryability, or what failures mean. The descriptions are under 100 chars and lack context on when to use each tool or what they return. This is a thin wrapper with no agent-optimization.

• Document the output schema for list_tools: return type should be an array of tool objects with 'name', 'description', and 'inputSchema' fields so LLMs can discover upstream capabilities.
• Replace the bare 'arguments' object in call_tool with a description explaining that it must match the upstream tool's input schema. Consider adding a 'tool_schema' parameter that returns the schema for validation before calling.
• Add 10 - 20 char descriptions to each parameter. E.g., 'name: The exact name of the upstream Habitify tool to invoke (e.g., "create_habit", "log_completion").'
• Document call_tool's return value: 'Returns the result from the upstream tool, or an error object with code, message, and recovery hint if the tool fails.'
• Add error handling: 'If the tool name is invalid, call list_tools first to discover available tools. If auth fails, check that HABITIFY_REFRESH_TOKEN is set and valid.'

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

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