FastAPI-based gateway that bridges mobile AI applications with MCP servers for Meraki and Catalyst network platforms, providing device inventory, health monitoring, compliance data, and AI-powered insights via AWS Bedrock.
Both tools have minimal descriptions (under 100 chars) and critically, NO visible input schemas in the source code. The tools are registered in api_gateway.py but their parameter schemas are not shown. Tool names follow verb_noun convention (get_*), which is positive, but lack of schemas is a hard blocker. Descriptions are generic and do not explain WHEN to use each tool or what they return. No error handling guidance, no output schema documentation, and no parameter descriptions visible.
Retrieves configured Catalyst clusters from the Catalyst MCP server
Retrieves configured Meraki organizations from the Meraki MCP server
Tool descriptions are under 100 characters and lack context. Missing: WHEN to use, WHAT it returns, any prerequisites or dependencies.
No output schema documented. LLMs cannot plan downstream calls or extract required fields (e.g., organization IDs, cluster IDs) without knowing response structure.
No error handling guidance. If a tool fails (e.g., MCP server unreachable), the LLM has no recovery path or retry strategy.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 35 | <=2025-11-25 | v2 |
Tool definitions appear to be inferred from code rather than explicitly registered with full schemas.