Loading ToolBench
Network Intent Validation Tool — read-only MCP server for validating live network state against network intent
Critical gaps across all 8 tools. No input schemas visible in source code, all tools score 0 for schema. Descriptions are present but generic (avg ~60 chars, below 194-char baseline). No parameter documentation. Tool names follow verb_noun pattern (positive), but lack specificity for distinguishing similar protocol tools (get_ospf, get_bgp, get_eigrp are semantically similar). No output schema documentation. No error handling guidance. No security annotations. Tools are read-only (low risk), but definitions do not meet production standards.
Retrieve BGP protocol state and configuration from network devices
Retrieve EIGRP protocol state and configuration from network devices
Retrieve network intent configuration from NetBox config contexts
Retrieve interface state and configuration from network devices
Retrieve OSPF protocol state and configuration from network devices
Retrieve routing table and route information from network devices
No input schemas visible in source code. All 8 tools lack documented parameter types, constraints, and descriptions. LLMs cannot infer valid parameter values or ranges.
Tool descriptions are generic and under 50 characters. No explanation of WHEN to use each tool, WHAT it returns, or HOW it differs from similar tools (e.g., get_ospf vs get_bgp vs get_eigrp).
No parameter descriptions. Tools accept parameters (implied by function signatures in tools/ modules) but no documentation visible for what each parameter controls, its type, or valid values.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 40 | <=2025-11-25 | v2 |
Retrieve routing policy and route-map configuration from network devices
Execute arbitrary show commands on network devices
No output schema documentation. LLMs cannot plan downstream calls or extract required fields (e.g., device_id, protocol_state) without knowing response structure.
No error handling guidance. Tools do not document what errors can occur, whether they are retryable, or what the LLM should do next (e.g., 'Device unreachable, try a different device or check connectivity').