Loading ToolBench
MCP server for optimizing web-to-print factory capacity allocation and aggregating print jobs across partner printing factories
Single tool 'execute_skill_action' has a reasonable name starting with a verb, but critical gaps undermine quality. Description is 89 chars (below 10-char minimum for actionable guidance). Input schema is present with 3 parameters, all typed as string/integer, but parameter descriptions are minimal and lack constraints (e.g., no enum for print_type, no range for run_quantity/deadline_days). Output schema is completely undocumented, the response structure is inferred from client.py but not formally declared in the tool definition. No error handling guidance, no idempotency markers, no security considerations. The tool modifies state (assigns factories, generates costs) but the description does not indicate this is a command tool with side effects.
Execute factory capacity aggregation and print job optimization
Output schema completely undocumented. Response fields (print_job_id, assigned_partner_printing_factory, idle_factory_capacity_utilized_pct, etc.) are inferred from client.py but not formally declared in tool definition. LLMs cannot plan downstream operations without knowing what fields to expect.
Tool description (89 chars) lacks actionable context. Does not explain WHEN to use this tool, WHAT it optimizes for, or that it modifies state (assigns factories, generates costs). Descriptions should be 50-200 chars and answer: what does it do, when should the LLM call it, what does it return?
Parameter 'print_type' accepts free-form strings with no enum constraint. Description mentions examples ('offset_color_flyers_a4', 'business_cards_matte_finish') but does not enumerate valid values. LLMs will hallucinate invalid print types.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 34 | 2026-07-28+ | v2 |
Numeric parameters 'run_quantity' and 'deadline_days' lack min/max bounds. No description of valid ranges. LLMs may pass absurd values (0 units, 365 days) that break business logic.
No error handling guidance. Tool description does not explain what happens if print_type is invalid, run_quantity exceeds factory capacity, or deadline is impossible. LLMs have no recovery path on failure.