linshenkx/prompt-optimizer · error · EvaluationParseError

Evaluation result is not a valid object.

Error message

Evaluation result is not a valid object.

What it means

EvaluationParseError thrown during normalizeEvaluationResponse when the parsed evaluation payload is not a plain object (null, array, primitive, or undefined). This guards the normalization step that expects an object containing at least a score. It indicates the extracted JSON, while parseable, is not an evaluation result object.

Source

Thrown at packages/core/src/services/evaluation/service.ts:3113

        if (depth === 0) {
          return content.slice(start, i + 1);
        }
      }
    }

    return null;
  }

  /**
   * 标准化评估响应（统一结构）
   */
  private normalizeEvaluationResponse(
    data: any,
    type: EvaluationType,
    metadata?: EvaluationResponse['metadata']
  ): EvaluationResponse {
    if (!data || typeof data !== 'object') {
      throw new EvaluationParseError('Evaluation result is not a valid object.');
    }

    if (data.score === undefined || data.score === null) {
      throw new EvaluationParseError('Evaluation result is missing the "score" field.');
    }

    // 提取分数（0-100，整数）
    const extractScore = (value: any, fieldName: string): number => {
      if (value === undefined || value === null) {
        throw new EvaluationParseError(`Evaluation result is missing score for "${fieldName}".`);
      }
      const num = typeof value === 'number' ? value : parseInt(String(value));
      if (isNaN(num)) {
        throw new EvaluationParseError(`Invalid numeric score for "${fieldName}": ${value}`);
      }
      return Math.max(0, Math.min(100, num));
    };

View on GitHub (pinned to 3e677b1d9f)

Solutions

1. Log the raw content and the extracted data to see the actual shape returned
2. Adjust the judge prompt to return a full JSON object with a score field
3. Wrap in retry with stricter instructions or a different judge model

if (typeof parsed !== 'object' || parsed === null || Array.isArray(parsed)) skipOrFlag(rawContent);

function isScoreObject(v: unknown): v is Record<string, unknown> { return typeof v === 'object' && v !== null && !Array.isArray(v); }

try { ... } catch (e) { if (e instanceof EvaluationParseError) { /* log raw content, adjust prompt, retry once */ } else throw e; }

When it happens

Trigger: The judge model returns a bare number/string (e.g. just "85") or an array, and it is fed into normalization; or JSON extraction grabs a non-object fragment from the output.

Common situations: Models answering with only the score; extraction regex capturing a JSON scalar; prompt template changed so the model returns a different top-level shape.

Related errors

• Failed to parse evaluation result: no valid score JSON or re
• Evaluation result is missing the "score" field.
• Evaluation result is missing score for "${fieldName}".
• Invalid numeric score for "${fieldName}": ${value}
• Evaluation result is missing a valid overall score.

AI-assisted analysis of linshenkx/prompt-optimizer@3e677b1d9f (2026-08-27). Data as JSON: /api/errors/71a6918ae47479c6. Report an issue: GitHub.