api.strale.io

{
  "example": {
    "changes_made": [
      {
        "change": "Added explicit role definition and system context",
        "reasoning": "Clarifies the purpose and expected behavior, reducing ambiguity about what the prompt should accomplish"
      },
      {
        "change": "Restructured vague input into clear task with specific output requirements",
        "reasoning": "Provides measurable evaluation criteria rather than implicit expectations"
      },
      {
        "change": "Added numbered output format expectations",
        "reasoning": "Enables consistent, structured responses that can be programmatically evaluated"
      },
      {
        "change": "Included meta-analysis request (confidence, limitations)",
        "reasoning": "Makes the testing process more comprehensive and diagnostic for capability assessment"
      }
    ],
    "clarity_score": {
      "improved": 9,
      "original": 1
    },
    "improved_prompt": "You are an automated capability testing system. Process the following test input and provide structured feedback on your ability to handle it: 'test_value'. Return your analysis including: (1) successful parsing of the input, (2) any constraints or limitations encountered, (3) confidence level in handling this input type, and (4) recommendations for improvement.",
    "improved_token_estimate": 65,
    "issues_found": [
      "Extremely vague - unclear what should happen with 'test_value'",
      "No explicit task definition or expected output format",
      "Cannot be effectively evaluated or reproduced",
      "No context for what 'automated capability testing' means",
      "Insufficient for meaningful testing purposes"
    ],
    "original_token_estimate": 2,
    "techniques_applied": [
      "Role-based prompting",
      "Task decomposition with numbered steps",
      "Explicit output structure definition",
      "Context clarification",
      "Meta-cognitive prompting for self-assessment"
    ]
  },
  "properties": {
    "changes_made": {
      "type": "array"
    },
    "clarity_score": {
      "type": "object"
    },
    "improved_prompt": {
      "type": "string"
    },
    "techniques_applied": {
      "type": "array"
    }
  },
  "type": "object"
}