Loading ToolBench
Local-first visual generation runtime and studio for people and coding agents, with reproducible image and video workflows across multiple providers.
Critical gaps across all dimensions. Tool definitions are inferred from agentToolFormatting.ts without visible explicit schema registration, parameter types, or descriptions in the source. No input schemas found in provided code. Descriptions are minimal (10-50 chars) and lack LLM-optimized guidance. No parameter documentation visible. No output schema definitions. Error handling absent. This server cannot be reliably used by agents without substantial schema work.
Generate images using configured providers
Generate videos using configured providers
Get generation errors for debugging
Get image context for agent tool calls
Perform web search for agent tool calls
No input schemas visible in source code. Tool definitions appear inferred from file references only; explicit schema registration with JSON Schema types and constraints not found.
Descriptions are too short (10-50 chars) and lack LLM-optimized guidance. Missing WHAT the tool does, WHEN to use it, and what it returns. Rubric baseline: 194 chars average, 10-1024 range.
No parameter documentation visible. Cannot verify parameter types, descriptions, enums, or constraints. Rubric requires every parameter to have a description explaining what it controls.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 35 | 2026-07-28+ | v2 |
No output schema documentation. LLMs cannot plan downstream tool calls or extract required data without knowing response structure and field types.
No error handling guidance. Tools that fail (e.g., image generation timeout, API rate limit) provide no recovery hints or error classification (retryable vs. fatal).