📎 feat: Upload as Text Support for Plaintext, STT, RAG, and Token Limits (#8868)

* 🪶 feat: Add Support for Uploading Plaintext Files

feat: delineate between OCR and text handling in fileConfig field of config file

- also adds support for passing in mimetypes as just plain file extensions

feat: add showLabel bool to support future synthetic component DynamicDropdownInput

feat: add new combination dropdown-input component in params panel to support file type token limits

refactor: move hovercard to side to align with other hovercards

chore: clean up autogenerated comments

feat: add delineation to file upload path between text and ocr configured filetypes

feat: add token limit checks during file upload

refactor: move textParsing out of ocrEnabled logic

refactor: clean up types for filetype config

refactor: finish decoupling DynamicDropdownInput from fileTokenLimits

fix: move image token cost function into file to fix circular dependency causing unittest to fail and remove unused var for linter

chore: remove out of scope code following review

refactor: make fileTokenLimit conform to existing styles

chore: remove unused localization string

chore: undo changes to DynamicInput and other strays

feat: add fileTokenLimit to all provider config panels

fix: move textParsing back into ocr tool_resource block for now so that it doesn't interfere with other upload types

* 📤 feat: Add RAG API Endpoint Support for Text Parsing (#8849)

* feat: implement RAG API integration for text parsing with fallback to native parsing

* chore: remove TODO now that placeholder and fllback are implemented

* ✈️ refactor: Migrate Text Parsing to TS (#8892)

* refactor: move generateShortLivedToken to packages/api

* refactor: move textParsing logic into packages/api

* refactor: reduce nesting and dry code with createTextFile

* fix: add proper source handling

* fix: mock new parseText and parseTextNative functions in jest file

* ci: add test coverage for textParser

* 💬 feat: Add Audio File Support to Upload as Text (#8893)

* feat: add STT support for Upload as Text

* refactor: move processAudioFile to packages/api

* refactor: move textParsing from utils to files

* fix: remove audio/mp3 from unsupported mimetypes test since it is now supported

* ✂️ feat: Configurable File Token Limits and Truncation (#8911)

* feat: add configurable fileTokenLimit default value

* fix: add stt to fileConfig merge logic

* fix: add fileTokenLimit to mergeFileConfig logic so configurable value is actually respected from yaml

* feat: add token limiting to parsed text files

* fix: add extraction logic and update tests so fileTokenLimit isnt sent to LLM providers

* fix: address comments

* refactor: rename textTokenLimiter.ts to text.ts

* chore: update form-data package to address CVE-2025-7783 and update package-lock

* feat: use default supported mime types for ocr on frontend file validation

* fix: should be using logger.debug not console.debug

* fix: mock existsSync in text.spec.ts

* fix: mock logger rather than every one of its function calls

* fix: reorganize imports and streamline file upload processing logic

* refactor: update createTextFile function to use destructured parameters and improve readability

* chore: update file validation to use EToolResources for improved type safety

* chore: update import path for types in audio processing module

* fix: update file configuration access and replace console.debug with logger.debug for improved logging

---------

Co-authored-by: Dustin Healy <dustinhealy1@gmail.com>
Co-authored-by: Dustin Healy <54083382+dustinhealy@users.noreply.github.com>

api/server/routes/files/files.js

@@ -387,7 +387,8 @@ router.post('/', async (req, res) => {
 
     if (
       error.message?.includes('Invalid file format') ||
-      error.message?.includes('No OCR result')
+      error.message?.includes('No OCR result') ||
+      error.message?.includes('exceeds token limit')
     ) {
       message = error.message;
     }

api/server/routes/files/images.js

@@ -29,6 +29,17 @@ router.post('/', async (req, res) => {
   } catch (error) {
     // TODO: delete remote file if it exists
     logger.error('[/files/images] Error processing file:', error);
+
+    let message = 'Error processing file';
+
+    if (
+      error.message?.includes('Invalid file format') ||
+      error.message?.includes('No OCR result') ||
+      error.message?.includes('exceeds token limit')
+    ) {
+      message = error.message;
+    }
+
     try {
       const filepath = path.join(
         appConfig.paths.imageOutput,
@@ -39,7 +50,7 @@ router.post('/', async (req, res) => {
     } catch (error) {
       logger.error('[/files/images] Error deleting file:', error);
     }
-    res.status(500).json({ message: 'Error processing file' });
+    res.status(500).json({ message });
   } finally {
     try {
       await fs.unlink(req.file.path);

api/server/services/AuthService.js

@@ -500,18 +500,6 @@ const resendVerificationEmail = async (req) => {
     };
   }
 };
-/**
- * Generate a short-lived JWT token
- * @param {String} userId - The ID of the user
- * @param {String} [expireIn='5m'] - The expiration time for the token (default is 5 minutes)
- * @returns {String} - The generated JWT token
- */
-const generateShortLivedToken = (userId, expireIn = '5m') => {
-  return jwt.sign({ id: userId }, process.env.JWT_SECRET, {
-    expiresIn: expireIn,
-    algorithm: 'HS256',
-  });
-};
 
 module.exports = {
   logoutUser,
@@ -522,5 +510,4 @@ module.exports = {
   setOpenIDAuthTokens,
   requestPasswordReset,
   resendVerificationEmail,
-  generateShortLivedToken,
 };

api/server/services/Endpoints/anthropic/build.js

@@ -6,6 +6,7 @@ const buildOptions = (endpoint, parsedBody) => {
     modelLabel,
     promptPrefix,
     maxContextTokens,
+    fileTokenLimit,
     resendFiles = anthropicSettings.resendFiles.default,
     promptCache = anthropicSettings.promptCache.default,
     thinking = anthropicSettings.thinking.default,
@@ -29,6 +30,7 @@ const buildOptions = (endpoint, parsedBody) => {
     greeting,
     spec,
     maxContextTokens,
+    fileTokenLimit,
     modelOptions,
   });
 

api/server/services/Endpoints/bedrock/build.js

@@ -6,6 +6,7 @@ const buildOptions = (endpoint, parsedBody) => {
     modelLabel: name,
     promptPrefix,
     maxContextTokens,
+    fileTokenLimit,
     resendFiles = true,
     imageDetail,
     iconURL,
@@ -24,6 +25,7 @@ const buildOptions = (endpoint, parsedBody) => {
     spec,
     promptPrefix,
     maxContextTokens,
+    fileTokenLimit,
     model_parameters,
   });
 

api/server/services/Endpoints/custom/build.js

@@ -7,6 +7,7 @@ const buildOptions = (endpoint, parsedBody, endpointType) => {
     chatGptLabel,
     promptPrefix,
     maxContextTokens,
+    fileTokenLimit,
     resendFiles = true,
     imageDetail,
     iconURL,
@@ -27,6 +28,7 @@ const buildOptions = (endpoint, parsedBody, endpointType) => {
     greeting,
     spec,
     maxContextTokens,
+    fileTokenLimit,
     modelOptions,
   });
 

api/server/services/Endpoints/google/build.js

@@ -12,6 +12,7 @@ const buildOptions = (endpoint, parsedBody) => {
     spec,
     artifacts,
     maxContextTokens,
+    fileTokenLimit,
     ...modelOptions
   } = parsedBody;
   const endpointOption = removeNullishValues({
@@ -24,6 +25,7 @@ const buildOptions = (endpoint, parsedBody) => {
     greeting,
     spec,
     maxContextTokens,
+    fileTokenLimit,
     modelOptions,
   });
 

api/server/services/Endpoints/openAI/build.js

@@ -7,6 +7,7 @@ const buildOptions = (endpoint, parsedBody) => {
     chatGptLabel,
     promptPrefix,
     maxContextTokens,
+    fileTokenLimit,
     resendFiles = true,
     imageDetail,
     iconURL,
@@ -27,6 +28,7 @@ const buildOptions = (endpoint, parsedBody) => {
     greeting,
     spec,
     maxContextTokens,
+    fileTokenLimit,
     modelOptions,
   });
 

api/server/services/Files/Audio/STTService.js

@@ -319,4 +319,4 @@ async function speechToText(req, res) {
   await sttService.processSpeechToText(req, res);
 }
 
-module.exports = { speechToText };
+module.exports = { STTService, speechToText };

api/server/services/Files/Local/crud.js

@@ -3,7 +3,7 @@ const path = require('path');
 const axios = require('axios');
 const { logger } = require('@librechat/data-schemas');
 const { EModelEndpoint } = require('librechat-data-provider');
-const { generateShortLivedToken } = require('~/server/services/AuthService');
+const { generateShortLivedToken } = require('@librechat/api');
 const { getBufferMetadata } = require('~/server/utils');
 const paths = require('~/config/paths');
 

api/server/services/Files/VectorDB/crud.js

@@ -1,10 +1,9 @@
 const fs = require('fs');
 const axios = require('axios');
 const FormData = require('form-data');
-const { logAxiosError } = require('@librechat/api');
 const { logger } = require('@librechat/data-schemas');
 const { FileSources } = require('librechat-data-provider');
-const { generateShortLivedToken } = require('~/server/services/AuthService');
+const { logAxiosError, generateShortLivedToken } = require('@librechat/api');
 
 /**
  * Deletes a file from the vector database. This function takes a file object, constructs the full path, and

api/server/services/Files/images/encode.js

@@ -1,13 +1,16 @@
 const axios = require('axios');
-const { logAxiosError } = require('@librechat/api');
+const { logger } = require('@librechat/data-schemas');
+const { logAxiosError, processTextWithTokenLimit } = require('@librechat/api');
 const {
   FileSources,
   VisionModes,
   ImageDetail,
   ContentTypes,
   EModelEndpoint,
+  mergeFileConfig,
 } = require('librechat-data-provider');
 const { getStrategyFunctions } = require('~/server/services/Files/strategies');
+const countTokens = require('~/server/utils/countTokens');
 
 /**
  * Converts a readable stream to a base64 encoded string.
@@ -102,11 +105,28 @@ async function encodeAndFormat(req, files, endpoint, mode) {
     return result;
   }
 
+  const fileTokenLimit =
+    req.body?.fileTokenLimit ?? mergeFileConfig(req.config?.fileConfig).fileTokenLimit;
+
   for (let file of files) {
     /** @type {FileSources} */
     const source = file.source ?? FileSources.local;
     if (source === FileSources.text && file.text) {
-      result.text += `${!result.text ? 'Attached document(s):\n```md' : '\n\n---\n\n'}# "${file.filename}"\n${file.text}\n`;
+      let fileText = file.text;
+
+      const { text: limitedText, wasTruncated } = await processTextWithTokenLimit({
+        text: fileText,
+        tokenLimit: fileTokenLimit,
+        tokenCountFn: (text) => countTokens(text),
+      });
+
+      if (wasTruncated) {
+        logger.debug(
+          `[encodeAndFormat] Text content truncated for file: ${file.filename} due to token limits`,
+        );
+      }
+
+      result.text += `${!result.text ? 'Attached document(s):\n```md' : '\n\n---\n\n'}# "${file.filename}"\n${limitedText}\n`;
     }
 
     if (!file.height) {
@@ -135,7 +155,7 @@ async function encodeAndFormat(req, files, endpoint, mode) {
         base64Data = null;
         continue;
       } catch (error) {
-        // Error handling code
+        logger.error('Error processing image from blob storage:', error);
       }
     } else if (source !== FileSources.local && base64Only.has(endpoint)) {
       const [_file, imageURL] = await preparePayload(req, file);

api/server/services/Files/process.js

@@ -17,7 +17,8 @@ const {
   isAssistantsEndpoint,
 } = require('librechat-data-provider');
 const { EnvVar } = require('@librechat/agents');
-const { sanitizeFilename } = require('@librechat/api');
+const { logger } = require('@librechat/data-schemas');
+const { sanitizeFilename, parseText, processAudioFile } = require('@librechat/api');
 const {
   convertImage,
   resizeAndConvert,
@@ -33,7 +34,7 @@ const { checkCapability } = require('~/server/services/Config');
 const { LB_QueueAsyncCall } = require('~/server/utils/queue');
 const { getStrategyFunctions } = require('./strategies');
 const { determineFileType } = require('~/server/utils');
-const { logger } = require('~/config');
+const { STTService } = require('./Audio/STTService');
 
 /**
  * Creates a modular file upload wrapper that ensures filename sanitization
@@ -552,51 +553,84 @@ const processAgentFileUpload = async ({ req, res, metadata }) => {
     }
     // Note: File search processing continues to dual storage logic below
   } else if (tool_resource === EToolResources.ocr) {
-    const isOCREnabled = await checkCapability(req, AgentCapabilities.ocr);
-    if (!isOCREnabled) {
-      throw new Error('OCR capability is not enabled for Agents');
-    }
-
-    const { handleFileUpload: uploadOCR } = getStrategyFunctions(
-      appConfig?.ocr?.strategy ?? FileSources.mistral_ocr,
-    );
     const { file_id, temp_file_id = null } = metadata;
 
-    const {
-      text,
-      bytes,
-      // TODO: OCR images support?
-      images: _i,
-      filename,
-      filepath: ocrFileURL,
-    } = await uploadOCR({ req, appConfig, file, loadAuthValues });
-
-    const fileInfo = removeNullishValues({
-      text,
-      bytes,
-      file_id,
-      temp_file_id,
-      user: req.user.id,
-      type: 'text/plain',
-      filepath: ocrFileURL,
-      source: FileSources.text,
-      filename: filename ?? file.originalname,
-      model: messageAttachment ? undefined : req.body.model,
-      context: messageAttachment ? FileContext.message_attachment : FileContext.agents,
-    });
-
-    if (!messageAttachment && tool_resource) {
-      await addAgentResourceFile({
-        req,
+    /**
+     * @param {object} params
+     * @param {string} params.text
+     * @param {number} params.bytes
+     * @param {string} params.filepath
+     * @param {string} params.type
+     * @return {Promise<void>}
+     */
+    const createTextFile = async ({ text, bytes, filepath, type = 'text/plain' }) => {
+      const fileInfo = removeNullishValues({
+        text,
+        bytes,
         file_id,
-        agent_id,
-        tool_resource,
+        temp_file_id,
+        user: req.user.id,
+        type,
+        filepath: filepath ?? file.path,
+        source: FileSources.text,
+        filename: file.originalname,
+        model: messageAttachment ? undefined : req.body.model,
+        context: messageAttachment ? FileContext.message_attachment : FileContext.agents,
       });
+
+      if (!messageAttachment && tool_resource) {
+        await addAgentResourceFile({
+          req,
+          file_id,
+          agent_id,
+          tool_resource,
+        });
+      }
+      const result = await createFile(fileInfo, true);
+      return res
+        .status(200)
+        .json({ message: 'Agent file uploaded and processed successfully', ...result });
+    };
+
+    const fileConfig = mergeFileConfig(appConfig.fileConfig);
+
+    const shouldUseOCR = fileConfig.checkType(
+      file.mimetype,
+      fileConfig.ocr?.supportedMimeTypes || [],
+    );
+
+    if (shouldUseOCR && !(await checkCapability(req, AgentCapabilities.ocr))) {
+      throw new Error('OCR capability is not enabled for Agents');
+    } else if (shouldUseOCR) {
+      const { handleFileUpload: uploadOCR } = getStrategyFunctions(
+        appConfig?.ocr?.strategy ?? FileSources.mistral_ocr,
+      );
+      const { text, bytes, filepath: ocrFileURL } = await uploadOCR({ req, file, loadAuthValues });
+      return await createTextFile({ text, bytes, filepath: ocrFileURL });
     }
-    const result = await createFile(fileInfo, true);
-    return res
-      .status(200)
-      .json({ message: 'Agent file uploaded and processed successfully', ...result });
+
+    const shouldUseSTT = fileConfig.checkType(
+      file.mimetype,
+      fileConfig.stt?.supportedMimeTypes || [],
+    );
+
+    if (shouldUseSTT) {
+      const sttService = await STTService.getInstance();
+      const { text, bytes } = await processAudioFile({ file, sttService });
+      return await createTextFile({ text, bytes });
+    }
+
+    const shouldUseText = fileConfig.checkType(
+      file.mimetype,
+      fileConfig.text?.supportedMimeTypes || [],
+    );
+
+    if (!shouldUseText) {
+      throw new Error(`File type ${file.mimetype} is not supported for OCR or text parsing`);
+    }
+
+    const { text, bytes } = await parseText({ req, file, file_id });
+    return await createTextFile({ text, bytes, type: file.mimetype });
   }
 
   // Dual storage pattern for RAG files: Storage + Vector DB

api/server/services/Files/processFiles.test.js

@@ -101,6 +101,11 @@ jest.mock('~/server/utils', () => ({
   determineFileType: jest.fn(),
 }));
 
+jest.mock('@librechat/api', () => ({
+  parseText: jest.fn(),
+  parseTextNative: jest.fn(),
+}));
+
 // Import the actual processFiles function after all mocks are set up
 const { processFiles } = require('./process');
 const { updateFileUsage } = require('~/models/File');