MCP Character Tools

Instructions

Implementation Reference

• src/tools/counting.ts:27-53 (handler)

  Core implementation of the count_letter tool logic: normalizes case if needed, finds all positions of the letter via loop, computes visual breakdown and density summary using helpers, returns structured output.

  export function countLetter(input: CountLetterInput): CountLetterOutput {
    const { text, letter, case_sensitive } = input;
    
    // Normalize the letter to search for
    const searchLetter = case_sensitive ? letter : letter.toLowerCase();
    const searchText = case_sensitive ? text : text.toLowerCase();
    
    const positions: number[] = [];
    for (let i = 0; i < searchText.length; i++) {
      if (searchText[i] === searchLetter) {
        positions.push(i);
      }
    }
  
    const visual = createVisualBreakdown(text, positions, letter);
    const density = createDensitySummary(positions.length, text.length, letter);
  
    return {
      text,
      letter,
      count: positions.length,
      case_sensitive,
      positions,
      visual,
      density,
    };
  }

• src/tools/counting.ts:11-25 (schema)

  TypeScript interfaces defining the input schema (text, letter, case_sensitive) and output schema (including count, positions, visual, density) for the countLetter function.

  export interface CountLetterInput {
    text: string;
    letter: string;
    case_sensitive: boolean;
  }
  
  export interface CountLetterOutput {
    text: string;
    letter: string;
    count: number;
    case_sensitive: boolean;
    positions: number[];
    visual: string;
    density: string;
  }

• src/index.ts:52-88 (registration)

  MCP server registration of the 'count_letter' tool, including title, description, Zod input schema validation, annotations, and async handler that invokes countLetter and formats MCP response with density, visual, count, and positions.

  server.registerTool(
    "count_letter",
    {
      title: "Count Letter",
      description: `Count occurrences of a specific letter in text.
  
  Returns the count, positions, a visual breakdown showing where each letter appears, and a density percentage.
  
  Args:
    - text (string): The text to analyze
    - letter (string): The single letter to count
    - case_sensitive (boolean): Whether to match case exactly (default: false)
  
  Returns: count, positions array, visual breakdown, and density summary.
  
  Example: count_letter("strawberry", "r") → count: 3, positions: [2, 5, 8]`,
      inputSchema: z.object({
        text: z.string().min(1).describe("The text to analyze"),
        letter: z.string().length(1).describe("The letter to count"),
        case_sensitive: z.boolean().default(false).describe("Match case exactly"),
      }).strict(),
      annotations: { readOnlyHint: true, destructiveHint: false, idempotentHint: true, openWorldHint: false },
    },
    async (params) => {
      const result = countLetter({
        text: params.text,
        letter: params.letter,
        case_sensitive: params.case_sensitive,
      });
      return {
        content: [{
          type: "text" as const,
          text: `${result.density}\n\n${result.visual}\n\nResult: ${JSON.stringify({ count: result.count, positions: result.positions })}`
        }],
      };
    }
  );