UNO-MCP

Instructions

Implementation Reference

• src/index.js:78-207 (handler)

  The primary handler function for the 'analyze_text' tool. It performs detailed text analysis including statistics, scene type detection, mood analysis, repetition detection, and generates a comprehensive markdown-formatted report.

  function analyzeText(text) {
    const wordCount = text.split(/\s+/).filter(Boolean).length;
    const charCount = text.length;
    
    // Basic analysis calculations
    const averageSentenceLength = 
      wordCount / (text.match(/[.!?]+/g)?.length || 1);
    
    // Simple repetition detection (counts words that appear more than once)
    const words = text.toLowerCase().match(/\b\w+\b/g) || [];
    const wordFreq = {};
    words.forEach(word => {
      wordFreq[word] = (wordFreq[word] || 0) + 1;
    });
    
    const repeatedWords = Object.entries(wordFreq)
      .filter(([_, count]) => count > 1)
      .sort((a, b) => b[1] - a[1])
      .slice(0, 5)
      .map(([word, count]) => `"${word}" (${count} times)`);
    
    // Detect scene type
    let sceneType = 'exposition';
    if (text.includes('!') || text.match(/run|jump|fight|chase|grab|throw|dash|leap|slam/gi)) {
      sceneType = 'action';
    } else if ((text.match(/"|'|said|asked|replied|whispered|shouted/g) || []).length > wordCount/20) {
      sceneType = 'dialogue';
    }
    
    // Detect mood
    let mood = 'neutral';
    const positiveWords = text.match(/happy|joy|smile|laugh|wonderful|beautiful|success|love|hope/gi) || [];
    const negativeWords = text.match(/sad|fear|angry|hate|despair|pain|suffer|terrible|awful/gi) || [];
    
    if (positiveWords.length > negativeWords.length) {
      mood = 'positive';
    } else if (negativeWords.length > positiveWords.length) {
      mood = 'negative';
    }
    
    // Create an analysis report in markdown format
    return `UNO Analysis Report
  
  
  ## Text Statistics
  - **Original Word Count**: ${wordCount}
  - **Original Character Count**: ${charCount}
  - **Target Word Count (200%)**: ${wordCount * 2}
  - **Target Character Count (200%)**: ${charCount * 2}
  
  ## Contextual Assessment
  
  ### Narrative Position
  - **Position**: ${text.length < 200 ? 'beginning' : (text.length > 1000 ? 'climax/turning point' : 'middle')}
  - **Introduction Markers**: ${text.length < 300 ? 'Present' : 'Absent'}
  - **Climax Markers**: ${text.match(/suddenly|realized|moment|revelation|truth/gi) ? 'Present' : 'Absent'}
  - **Resolution Markers**: ${text.match(/finally|end|last|resolution|conclusion/gi) ? 'Present' : 'Absent'}
  
  ### Character Focus
  - **Point of View**: ${text.match(/\bI\b|\bme\b|\bmy\b/gi) ? 'first-person' : 'third-person'}
  - **Potential Characters**: ${Array.from(new Set(text.match(/[A-Z][a-z]+/g) || [])).slice(0, 7).join(' ')}
  - **Pronoun Distribution**:
    - First Person: ${(text.match(/\bI\b|\bme\b|\bmy\b|\bmine\b|\bmyself\b/gi) || []).length}
    - Second Person: ${(text.match(/\byou\b|\byour\b|\byours\b|\byourself\b/gi) || []).length}
    - Third Person: ${(text.match(/\bhe\b|\bshe\b|\bit\b|\bhim\b|\bher\b|\bhis\b|\bits\b|\bthey\b|\bthem\b|\btheir\b/gi) || []).length}
  
  ### Scene Type
  - **Dominant Type**: ${sceneType}
  - **Action Elements**: ${sceneType === 'action' ? 'Strong' : 'Weak'}
  - **Dialogue Elements**: ${sceneType === 'dialogue' ? 'Strong' : 'Weak'}
  - **Exposition Elements**: ${sceneType === 'exposition' ? 'Strong' : 'Weak'}
  
  ### Mood and Tone
  - **Dominant Mood**: ${mood}
  - **Positive Elements**: ${positiveWords.length > 0 ? 'Present' : 'Absent'}
  - **Negative Elements**: ${negativeWords.length > 0 ? 'Present' : 'Absent'}
  - **Suspense Elements**: ${text.match(/mystery|unknown|wonder|question|curious|suspense/gi) ? 'Present' : 'Absent'}
  
  ## Enhancement Recommendations
  
  ### Golden Shadow Enhancement
  - **Need Level**: ${wordCount < 100 ? 'high' : (wordCount < 300 ? 'medium' : 'low')}
  - **Underdeveloped Characters**: ${Array.from(new Set(text.match(/[A-Z][a-z]+/g) || [])).slice(0, 4).join(' ')}
  - **Underdeveloped Plot Elements**: ${text.length < 500 ? 'Present' : 'Absent'}
  
  ### Environmental Expansion
  - **Need Level**: ${(text.match(/saw|looked|appeared|seemed|felt|heard|smelled|tasted/gi) || []).length < wordCount/20 ? 'high' : 'low'}
  - **Sensory Richness (0-4)**: ${Math.min(4, Math.floor((text.match(/saw|looked|appeared|seemed|felt|heard|smelled|tasted/gi) || []).length / (wordCount/100)))}
  - **Setting Description**: ${text.match(/room|house|building|street|city|forest|mountain|ocean|sky/gi) ? 'Present' : 'Absent'}
  - **Areas to Enhance**:
    - Visual: ${text.match(/saw|looked|appeared|seemed|watched|observed/gi) ? 'Sufficient' : 'Needs improvement'}
    - Auditory: ${text.match(/heard|sound|noise|voice|whisper|shout|bang|crash/gi) ? 'Sufficient' : 'Needs improvement'}
    - Tactile: ${text.match(/felt|touch|rough|smooth|hot|cold|soft|hard/gi) ? 'Sufficient' : 'Needs improvement'}
    - Olfactory: ${text.match(/smell|scent|odor|fragrance|stench/gi) ? 'Sufficient' : 'Needs improvement'}
  
  ### Action Scene Enhancement
  - **Applicability**: ${sceneType === 'action' ? 'Applicable' : 'Not applicable'}
  ${sceneType === 'action' ? `
  - **Current Intensity (0-4)**: ${Math.min(4, (text.match(/run|jump|fight|chase|grab|throw|dash|leap|slam/gi) || []).length)}
  - **Sensory Detail in Action**: ${(text.match(/saw|looked|appeared|seemed|felt|heard|smelled|tasted/gi) || []).length > (text.match(/run|jump|fight|chase|grab|throw|dash|leap|slam/gi) || []).length ? 'Sufficient' : 'Needs improvement'}
  - **Time Manipulation**: ${text.match(/suddenly|instant|moment|second|minute|time slowed|froze/gi) ? 'Present' : 'Absent'}
  - **Environmental Interaction**: ${text.match(/against|through|over|under|between|around/gi) ? 'Present' : 'Absent'}` : ''}
  
  ### Prose Smoothing
  - **Need Level**: ${averageSentenceLength > 20 || averageSentenceLength < 8 ? 'high' : 'low'}
  - **Average Sentence Length**: ${averageSentenceLength.toFixed(1)} words
  - **Sentence Length Variety**: ${(100 * (text.match(/\b\w{7,}\b/g)?.length || 0) / (text.match(/\b\w+\b/g)?.length || 1)).toFixed(1)}%
  - **Transition Words**: ${text.match(/however|therefore|meanwhile|subsequently|conversely|furthermore/gi) ? 'Present' : 'Absent'}
  - **Paragraph Length Variety**: ${text.split('\n\n').length > 1 ? 'Present' : 'Absent'}
  
  ### Repetition Elimination
  - **Severity**: ${repeatedWords.length > 10 ? 'high' : (repeatedWords.length > 5 ? 'medium' : 'low')}
  - **Repeated Word Count**: ${repeatedWords.length}
  - **Top Repeated Words**:
    - ${repeatedWords[0] || 'None'}
  ${repeatedWords[1] ? `  - ${repeatedWords[1]}` : ''}
  
  ## Repetition Patterns
  
  ### Repeated Words
  ${repeatedWords.slice(0, 4).map(word => `- ${word}`).join('\n') || '- No significant word repetition detected'}
  
  ### Repeated Phrases
  - No significant phrase repetition detected
  
  ### Repeated Sentence Structures
  - ${text.match(/[A-Z][^.!?]*\b(he|she|it|they)\b[^.!?]*/gi) ? 'pronoun-start pattern detected' : 'No significant structure repetition detected'}
  - ${text.match(/[A-Z][^.!?]*\b(was|were|had been)\b[^.!?]*/gi) ? 'passive-voice pattern detected' : ''}
  `;
  }

• src/index.js:12-21 (schema)

  Input schema definition for the 'analyze_text' tool, specifying the required 'text' parameter.

  const analyzeTextInputSchema = {
    type: 'object',
    properties: {
      text: {
        type: 'string',
        description: 'The story page text to analyze or enhance'
      }
    },
    required: ['text']
  };

• src/index.js:375-378 (registration)

  Registration of the 'analyze_text' tool in the list of available tools returned by ListToolsRequest.

    name: 'analyze_text',
    description: 'Analyzes text and generates a comprehensive report',
    inputSchema: analyzeTextInputSchema,
  },

• src/index.ts:164-179 (handler)

  TypeScript handler for 'analyze_text' tool that validates input and delegates to TextAnalyzer.analyzeText.

  private async handleAnalyzeText(args: any) {
    if (!args.text || typeof args.text !== 'string') {
      throw new McpError(ErrorCode.InvalidParams, 'Missing or invalid text parameter');
    }
  
    const report = await this.analyzer.analyzeText(args.text);
    
    return {
      content: [
        {
          type: 'text',
          text: report
        }
      ]
    };
  }

• src/index.ts:15-24 (schema)

  Input schema used for the 'analyze_text' tool in the TypeScript implementation.

  const textInputSchema = {
    type: 'object',
    properties: {
      text: {
        type: 'string',
        description: 'The story page text to analyze or enhance',
      }
    },
    required: ['text']
  };