[
  {
    "agent": "art-director",
    "issues": [
      "As a Prompt Engineering Auditor, I have evaluated the **art-director** agent prompt.\n\n### **Executive Summary**\nThe prompt is **highly sophisticated** and follows advanced architectural patterns (Role"
    ],
    "clarity": 5,
    "completeness": 5
  },
  {
    "agent": "copywriter",
    "clarity": 9,
    "completeness": 7,
    "actionability": 8,
    "issues": [
      "Lack of Output Formatting Standards: While the prompt defines what to write, it does not specify the structure of the final output (e.g., should it be a Markdown table, a JSON object, or a structured list?). This leads to inconsistent results.",
      "Missing 'Chain of Thought' Instructions: The agent is tasked with complex dual-role operations (copywriting + prompt engineering). There is no instruction to 'think' or 'analyze' the brief before generating, which may lead to superficial outputs.",
      "Undefined 'Input' Structure: The prompt mentions working from a 'Creative Director's brief' and 'Art Director's visual strategy,' but does not define the expected format of these inputs. This can lead to the agent hallucinating details if the input is sparse.",
      "Ambiguous Tool Usage: The tools listed (tag_element_intent, etc.) are presented as capabilities but lack instructions on *when* or *how* to invoke them. The agent may ignore them unless told to use them during the reasoning phase.",
      "No Quality Control/Feedback Loop: There is no instruction for the agent to self-correct or verify if the generated prompt meets the specific parameter syntax (e.g., checking for '--ar' in Midjourney) before finalizing."
    ],
    "suggestions": [
      "Define an Output Schema: Explicitly instruct the agent to return results in a specific format, such as: 'Headline: [text] | Subheadline: [text] | Midjourney Prompt: [prompt]'.",
      "Implement a 'Reasoning' Step: Add a section called 'Workflow' that instructs the agent to: 1. Analyze the brief using the provided tools, 2. Draft the copy, 3. Translate the visual strategy into technical prompts, 4. Final Review.",
      "Standardize Input Templates: Provide a template for the user to follow (e.g., 'Brief: [text], Visual Strategy: [text], Target Model: [model]') to ensure the agent always has the necessary context.",
      "Strengthen Tool Integration: Explicitly state: 'Before generating content, use `detect_mood` and `detect_topic` to establish the creative foundation.'",
      "Add a 'Constraints Checklist': Include a final instruction for the agent to verify its output against a checklist (e.g., 'Check: Is the headline < 8 words? Does the Midjourney prompt include aspect ratios?')."
    ]
  },
  {
    "agent": "graphic-designer",
    "issues": [
      "timeout"
    ],
    "clarity": 5,
    "completeness": 5
  },
  {
    "agent": "quality-inspector",
    "issues": [
      "timeout"
    ],
    "clarity": 5,
    "completeness": 5
  },
  {
    "agent": "brand-guardian",
    "issues": [
      "timeout"
    ],
    "clarity": 5,
    "completeness": 5
  },
  {
    "agent": "typographer",
    "issues": [
      "err: socket hang up"
    ],
    "clarity": 5,
    "completeness": 5
  },
  {
    "agent": "taste-scout",
    "issues": [
      "As a Prompt Engineering Auditor, I have reviewed the **\"taste-scout\"** prompt. \n\n### **Executive Summary**\nThe prompt is **high-performing and structurally sophisticated**. It moves beyond simple \"per"
    ],
    "clarity": 5,
    "completeness": 5
  },
  {
    "agent": "proofreader",
    "issues": [
      "timeout"
    ],
    "clarity": 5,
    "completeness": 5
  },
  {
    "agent": "editor",
    "issues": [
      "timeout"
    ],
    "clarity": 5,
    "completeness": 5
  },
  {
    "agent": "creative-director",
    "issues": [
      "timeout"
    ],
    "clarity": 5,
    "completeness": 5
  },
  {
    "agent": "foundation-architect",
    "issues": [
      "timeout"
    ],
    "clarity": 5,
    "completeness": 5
  }
]