feat: add KNOWLEDGE_ALLOW_PDF environment variable

README.md

@@ -65,6 +65,8 @@ The plugin can read almost any document:
 - **Documents:** `.pdf`, `.doc`, `.docx`
 - **Code Files:** `.js`, `.ts`, `.py`, `.java`, `.cpp`, `.html`, `.css` and many more
 
+> **Security Note:** PDF support can be disabled by setting `KNOWLEDGE_ALLOW_PDF=false` in your `.env` file if you prefer not to process PDF files.
+
 ## 💬 Using the Web Interface
 
 The plugin includes a web interface for managing documents!
@@ -137,8 +139,11 @@ OPENROUTER_API_KEY=your-openrouter-api-key
 ```env
 LOAD_DOCS_ON_STARTUP=true          # Auto-load from docs folder
 KNOWLEDGE_PATH=/custom/path        # Custom document path (default: ./docs)
+KNOWLEDGE_ALLOW_PDF=false          # Disable PDF support (default: true)
 ```
 
+**Note:** To disable PDF uploads for security reasons, set `KNOWLEDGE_ALLOW_PDF=false` in your `.env` file.
+
 ### Embedding Configuration
 
 ```env

package.json

@@ -1,7 +1,7 @@
 {
   "name": "@elizaos/plugin-knowledge",
   "description": "Plugin for Knowledge",
-  "version": "1.5.11",
+  "version": "1.5.12",
   "type": "module",
   "main": "dist/index.js",
   "module": "dist/index.js",

src/config.ts

@@ -2,12 +2,30 @@ import { ModelConfig, ModelConfigSchema, ProviderRateLimits } from './types.ts';
 import z from 'zod';
 import { logger, IAgentRuntime } from '@elizaos/core';
 
-const parseBooleanEnv = (value: any): boolean => {
+const parseBooleanEnv = (value: any, defaultValue: boolean = false): boolean => {
   if (typeof value === 'boolean') return value;
   if (typeof value === 'string') return value.toLowerCase() === 'true';
-  return false; // Default to false if undefined or other type
+  if (value === undefined || value === null) return defaultValue;
+  return defaultValue;
 };
 
+/**
+ * Checks if PDF support is enabled
+ * @param runtime The agent runtime to get settings from
+ * @returns true if PDF support is enabled, false otherwise
+ */
+export function isPdfAllowed(runtime?: IAgentRuntime): boolean {
+  const getSetting = (key: string, defaultValue?: string) => {
+    if (runtime) {
+      return runtime.getSetting(key) || process.env[key] || defaultValue;
+    }
+    return process.env[key] || defaultValue;
+  };
+
+  // Default to true (enabled) if not specified
+  return parseBooleanEnv(getSetting('KNOWLEDGE_ALLOW_PDF'), true);
+}
+
 /**
  * Validates the model configuration using runtime settings
  * @param runtime The agent runtime to get settings from

src/docs-loader.ts

@@ -4,6 +4,7 @@ import * as path from 'path';
 import { KnowledgeService } from './service.ts';
 import { AddKnowledgeOptions } from './types.ts';
 import { isBinaryContentType } from './utils.ts';
+import { isPdfAllowed } from './config.ts';
 
 /**
  * Get the knowledge path from runtime settings, environment, or default to ./docs
@@ -37,7 +38,8 @@ export async function loadDocsFromPath(
   service: KnowledgeService,
   agentId: UUID,
   worldId?: UUID,
-  knowledgePath?: string
+  knowledgePath?: string,
+  runtime?: any
 ): Promise<{ total: number; successful: number; failed: number }> {
   const docsPath = getKnowledgePath(knowledgePath);
 
@@ -48,6 +50,12 @@ export async function loadDocsFromPath(
 
   logger.info(`Loading documents from: ${docsPath}`);
 
+  // Check if PDF support is enabled
+  const pdfAllowed = isPdfAllowed(runtime);
+  if (!pdfAllowed) {
+    logger.warn('PDF support is disabled via KNOWLEDGE_ALLOW_PDF setting');
+  }
+
   // Get all files recursively
   const files = getAllFiles(docsPath);
 
@@ -80,6 +88,12 @@ export async function loadDocsFromPath(
         continue;
       }
 
+      // Skip PDF files if PDF support is disabled
+      if (contentType === 'application/pdf' && !pdfAllowed) {
+        logger.debug(`Skipping PDF file (PDF support disabled): ${filePath}`);
+        continue;
+      }
+
       // Read file
       const fileBuffer = fs.readFileSync(filePath);