All 7 tools have bilingual descriptions (Chinese/English) and basic input schemas with image_path parameters. However, descriptions are minimal (20-40 chars in English), lack actionable context for LLM selection, and do not explain when to use each tool vs. alternatives. Parameter descriptions are present but generic. No output schemas documented. No error handling guidance. Tool names are action-oriented (analyze_, extract_, describe_, diagnose_, understand_, code_from_) but lack clarity on distinctions, e.g., analyze_image vs. analyze_chart both analyze but differ in domain. Missing: pagination, structured output specs, recovery guidance, and per-parameter constraints.

• Expand each tool description to 100 - 200 characters. Include: (1) what the tool analyzes, (2) when to use it instead of similar tools, (3) expected output type. Example: 'Analyze a chart image to extract trends, values, and insights. Use this for bar charts, line graphs, pie charts, and heatmaps. Returns structured data: chart_type, axes, key_values, trends, insights.'
• Document output schemas for each tool. Define the response structure (e.g., { text: string, confidence?: number, structured_data?: object }). This enables LLMs to extract and chain results.
• Add parameter descriptions for 'prompt' in analyze_image. Explain: 'Optional custom question or instruction to guide the vision model. If omitted, uses the default analysis prompt. Max 500 chars.'
• Add error handling guidance to each tool description. Example: 'Returns error if image_path is invalid, file does not exist, or format is unsupported (JPEG, PNG, WebP, GIF). Suggest checking file path and format.'
• Clarify tool distinctions in descriptions. Rename or add suffixes to reduce ambiguity: analyze_image → analyze_general_image, understand_diagram → analyze_diagram, analyze_chart → analyze_data_chart. Or add explicit 'Use this for...' clauses in descriptions.