danny-avila/LibreChat · error · Error

Missing required field: prompt

Error message

Missing required field: prompt

What it means

Runtime guard at the top of the Gemini image tool's async callback: the destructured `prompt` argument must be present before initializeGeminiClient and the generation call proceed. Thrown before any Google API call.

Source

Thrown at api/app/clients/tools/structured/GeminiImageGen.js:326

 * Creates Gemini Image Generation tool
 * @param {Object} fields - Configuration fields
 * @returns {ReturnType<tool>} - The image generation tool
 */
function createGeminiImageTool(fields = {}) {
  const override = fields.override ?? false;

  if (!override && !fields.isAgent) {
    throw new Error('This tool is only available for agents.');
  }

  const { req, imageFiles = [], userId, fileStrategy, GEMINI_API_KEY, GOOGLE_KEY } = fields;

  const imageOutputType = fields.imageOutputType || EImageOutputType.PNG;

  const geminiImageGenTool = tool(
    async ({ prompt, image_ids, aspectRatio, imageSize }, runnableConfig) => {
      if (!prompt) {
        throw new Error('Missing required field: prompt');
      }

      logger.debug('[GeminiImageGen] Generating image', { aspectRatio, imageSize });

      let ai;
      try {
        ai = await initializeGeminiClient({
          GEMINI_API_KEY,
          GOOGLE_KEY,
        });
      } catch (error) {
        logger.error('[GeminiImageGen] Failed to initialize client:', error);
        return [
          [{ type: ContentTypes.TEXT, text: `Failed to initialize Gemini: ${error.message}` }],
          { content: [], file_ids: [] },
        ];
      }

View on GitHub (pinned to 5ff282f900)

Solutions

1. Ensure the tool-call arguments always include a non-empty `prompt` string.
2. Mark `prompt` as required in the tool's input schema so the model is forced to emit it.
3. Validate args before invoking when prompt comes from user input.
4. Log the raw tool arguments at the boundary to catch silent prompt drops.

Example fix

// before
await geminiImageTool.invoke({ image_ids: ['file-1'] }); // throws

// after
await geminiImageTool.invoke({ prompt: 'a fox under moonlight', image_ids: ['file-1'] });

function buildGeminiImageArgs(input) {
  if (typeof input.prompt !== 'string' || !input.prompt.trim()) {
    throw new Error('Gemini image tool requires a non-empty prompt.');
  }
  return input;
}

function hasPrompt(arg) {
  return typeof arg?.prompt === 'string' && arg.prompt.trim().length > 0;
}

try {
  await geminiTool.invoke(args);
} catch (e) {
  if (/Missing required field: prompt/.test(e.message)) return 'A prompt is required to generate an image.';
  throw e;
}

When it happens

Trigger: The Gemini image tool is invoked with arguments where `prompt` is missing, null, undefined, or empty — e.g., the model only passed image_ids or aspectRatio.

Common situations: LLM emitted a tool call with optional-only fields; schema change dropped prompt from required; programmatic invocation that built args from optional user input; prompt lost during arg marshalling upstream.

Related errors

• Missing required field: prompt
• Missing required field: prompt
• Missing required field: prompt
• Missing required field: finetune_id for finetuned generation
• Invalid endpoint for finetuned generation. Must be one of: $

AI-assisted analysis of danny-avila/LibreChat@5ff282f900 (2026-08-12). Data as JSON: /api/errors/5bd6d9cf0f08a791. Report an issue: GitHub.