Loading ToolBench
An MCP server that provides jokes about Disney animation movies and characters
Two tools with basic but incomplete definitions. Both have descriptions (good) and input schemas (good), but descriptions are generic and lack actionable context. Parameters lack type constraints (enums, ranges). No output schema documentation. Error handling is absent, tools return raw strings or dicts without guidance on failure modes. Naming is clear (verb_noun pattern). No security concerns (read-only operations, no secrets). Composition is simple but adequate for a joke generator.
Get a joke related to Disney animation movies and characters
Get a list of all available Disney joke categories
Parameter descriptions lack format constraints. 'character' and 'movie' accept free-form strings with no enum or pattern. LLMs will hallucinate invalid values like 'simba' (should be 'lion_king') or 'frozen_elsa' (should be 'frozen'). No guidance on valid options.
No output schema documentation. get_disney_joke returns a string; list_available_joke_categories returns a dict. LLMs cannot plan downstream operations or validate responses. Baseline: 100% of A+ tools document return types.
No error handling or recovery guidance. If a user requests a joke about 'barbie' (not in DISNEY_JOKES), the tool silently returns a generic joke. No error message, no hint to call list_available_joke_categories first. Agents cannot distinguish success from fallback.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 48 | 2026-07-28+ | v2 |
Tool descriptions are generic and lack WHEN/WHY context. 'Get a joke related to Disney animation movies and characters' does not explain when to use get_disney_joke vs list_available_joke_categories, or what the agent should do if a character is not found. Baseline: descriptions should answer what, when, and what to do next.
Parameter descriptions include example values ('mickey', 'elsa', 'frozen'). LLMs tend to reuse examples literally. Should replace with enum constraints or a reference to list_available_joke_categories.