A Model Context Protocol server for processing images using the Qwen 2.5 VL model to generate code or descriptions
Two tools with complete input schemas and basic descriptions, but significant quality gaps. Tool names lack action verbs (process_image_* is vague; should be generate_code_from_image, describe_image). Descriptions are minimal (20-30 chars, below the 50-200 char baseline for LLM optimization). Parameter descriptions are present but generic. No output schema documented. Error handling returns plain text errors without recovery guidance. No tool annotations (readOnlyHint, destructiveHint). The server reads images from URLs and calls an external API, but the interface does not guide LLM reasoning about when to use each tool or what to expect in responses.
处理图像并生成代码
处理图像并生成描述
Tool names lack action verbs. 'process_image_to_code' and 'process_image_to_description' are vague; LLMs cannot infer intent from 'process'. Should be 'generate_code_from_image' and 'describe_image' to match verb_noun pattern.
Descriptions are too short (20-30 chars). Baseline for LLM-optimized descriptions is 50-200 chars. Current descriptions ('处理图像并生成代码', '处理图像并生成描述') lack context on WHEN to use each tool, WHAT the output contains, and HOW it differs from the other tool.
No output schema documented. LLMs cannot plan downstream tool calls or extract structured data without knowing the response format. Tool returns {content: [{type: 'text', text: string}]} but this is not declared in the tool definition.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 45 | 2026-07-28+ | v2 |
Error handling returns plain text without recovery guidance. Errors like 'Download image failed: ...' or 'API response format incorrect' do not tell the LLM what to do next (retry, ask user, try different tool). Should follow pattern:recovery-guide.
No tool annotations. Both tools are read-only (marked as READ_ONLY in risk field) but lack readOnlyHint in schema. Tool annotations help LLMs reason about side effects and retry safety.