Loading ToolBench
Zero-dependency clients and portable agent resources for Zephior's public Discovery and MCP interfaces. Provides access to Zephior's site overview, resource search, pricing documents, and service recommendations.
Three tools with clear verb-noun naming and documented input schemas. All tools have descriptions (10-60 chars) and parameter types. However, descriptions are below the 50-200 char baseline for LLM optimization, parameter descriptions lack detail on constraints and formats, and output schemas are not documented. No error handling guidance visible. Tools are read-only and simple, but lack the depth needed for production agent reasoning.
Creates a pricing estimate for Ziva capacity credits with specified currency and billing period
Retrieves the Ziva pricing document for a specified currency
Recommends a Zephior service based on a requirement description
Descriptions are too brief (10-60 chars vs. 50-200 char baseline). LLMs cannot determine when to select these tools or how they differ from each other. 'Recommends a Zephior service based on a requirement description' lacks context on what 'service' means, when to use it vs. pricing tools, and what the output contains.
Parameter descriptions lack actionable constraints. 'Currency code: CHF, USD, or EUR' is present but 'Description of the service requirement' for recommend_zephior_service provides no guidance on length, format, or examples. LLMs cannot infer valid input ranges.
Output schemas are not documented. The TypeScript definitions show Promise<unknown> return types. Agents cannot plan downstream calls or extract required fields (e.g., does recommend_zephior_service return a service_id, name, price, or all three?).
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 63 | 2025-06-18+ | v2 |
No error handling guidance. The ZephiorPublicApiError class exists but tool descriptions do not explain what errors are possible, when they occur, or how to recover. An LLM receiving a 404 or rate-limit error has no guidance on next steps.
create_ziva_pricing_estimate parameter 'credits' has enum constraint in description but not in schema. Schema shows type:integer with no enum array. This forces LLMs to parse the description text rather than reading a machine-readable constraint.