Loading ToolBench
Installation and configuration package for OpenCode research assistant, including skills, agents, scripts, and MCP server configuration
Single tool 'screenshot' has a basic description (89 chars) and input schema with two parameters. However, the schema lacks proper JSON Schema type declarations (SavePath and OCR are described but not formally typed in the visible schema). The description is minimal and does not explain WHEN to use the tool, what it returns, or error handling. No output schema is documented. The tool name 'screenshot' lacks a verb prefix (should be 'capture_screenshot' or 'take_screenshot'). No error recovery guidance provided. This is a minimal, under-specified tool definition typical of early-stage implementations.
Capture full screen screenshot and optionally perform OCR using Tesseract to extract text from the image
Tool name lacks action verb. 'screenshot' should be 'capture_screenshot' or 'take_screenshot' to signal intent to LLMs.
Input schema lacks formal JSON Schema type declarations. Parameters SavePath and OCR are described but types are not explicitly declared in the visible schema structure.
No output schema documented. LLMs cannot plan downstream actions or extract returned data (file path, OCR text, success/failure status).
Description is minimal (89 chars) and lacks context. Does not explain WHEN to use the tool, what it returns, or how to handle OCR failures.
No error handling guidance. If Tesseract is not installed or OCR fails, the tool provides no recovery path for the LLM.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 33 | 2026-07-28+ | v2 |