Enable bypass of ip limitations via ENV in collector processing

.github/workflows/dev-build.yaml

@@ -6,7 +6,7 @@ concurrency:
 
 on:
   push:
-    branches: ['na'] # put your current branch to create a build. Core team only.
+    branches: ['3625-bypass-ip-check'] # put your current branch to create a build. Core team only.
     paths-ignore:
       - '**.md'
       - 'cloud-deployments/*'

collector/middleware/verifyIntegrity.js

@@ -1,9 +1,12 @@
 const { CommunicationKey } = require("../utils/comKey");
+const RuntimeSettings = require("../utils/runtimeSettings");
+const runtimeSettings = new RuntimeSettings();
 
 function verifyPayloadIntegrity(request, response, next) {
   const comKey = new CommunicationKey();
   if (process.env.NODE_ENV === "development") {
-    comKey.log('verifyPayloadIntegrity is skipped in development.')
+    comKey.log('verifyPayloadIntegrity is skipped in development.');
+    runtimeSettings.parseOptionsFromRequest(request);
     next();
     return;
   }
@@ -12,7 +15,9 @@ function verifyPayloadIntegrity(request, response, next) {
   if (!signature) return response.status(400).json({ msg: 'Failed integrity signature check.' })
 
   const validSignedPayload = comKey.verify(signature, request.body);
-  if (!validSignedPayload) return response.status(400).json({ msg: 'Failed integrity signature check.' })
+  if (!validSignedPayload) return response.status(400).json({ msg: 'Failed integrity signature check.' });
+
+  runtimeSettings.parseOptionsFromRequest(request);
   next();
 }
 

collector/utils/runtimeSettings/index.js

@@ -0,0 +1,83 @@
+const { reqBody } = require("../http");
+
+/**
+ * Runtime settings are used to configure the collector per-request.
+ * These settings are persisted across requests, but can be overridden per-request.
+ *
+ * The settings are passed in the request body via `options.runtimeSettings`
+ * which is set in the backend #attachOptions function in CollectorApi.
+ *
+ * We do this so that the collector and backend can share the same ENV variables
+ * but only pass the relevant settings to the collector per-request and be able to
+ * access them across the collector via a single instance of RuntimeSettings.
+ *
+ * TODO: We may want to set all options passed from backend to collector here,
+ * but for now - we are only setting the runtime settings specifically for backwards
+ * compatibility with existing CollectorApi usage.
+ */
+class RuntimeSettings {
+  static _instance = null;
+  settings = {};
+
+  // Any settings here will be persisted across requests
+  // and must be explicitly defined here.
+  settingConfigs = {
+    allowAnyIp: {
+      default: false,
+      // Value must be explicitly "true" or "false" as a string
+      validate: (value) => String(value) === "true",
+    },
+  };
+
+  constructor() {
+    if (RuntimeSettings._instance) return RuntimeSettings._instance;
+    RuntimeSettings._instance = this;
+    return this;
+  }
+
+  /**
+   * Parse the runtime settings from the request body options body
+   * see #attachOptions https://github.com/Mintplex-Labs/anything-llm/blob/ebf112007e0d579af3d2b43569db95bdfc59074b/server/utils/collectorApi/index.js#L18
+   * @param {import('express').Request} request
+   * @returns {void}
+   */
+  parseOptionsFromRequest(request = {}) {
+    const options = reqBody(request)?.options?.runtimeSettings || {};
+    for (const [key, value] of Object.entries(options)) {
+      if (!this.settingConfigs.hasOwnProperty(key)) continue;
+      this.set(key, value);
+    }
+    return;
+  }
+
+  /**
+   * Get a runtime setting
+   * - Will throw an error if the setting requested is not a supported runtime setting key
+   * - Will return the default value if the setting requested is not set at all
+   * @param {string} key
+   * @returns {any}
+   */
+  get(key) {
+    if (!this.settingConfigs[key])
+      throw new Error(`Invalid runtime setting: ${key}`);
+    return this.settings.hasOwnProperty(key)
+      ? this.settings[key]
+      : this.settingConfigs[key].default;
+  }
+
+  /**
+   * Set a runtime setting
+   * - Will throw an error if the setting requested is not a supported runtime setting key
+   * - Will validate the value against the setting's validate function
+   * @param {string} key
+   * @param {any} value
+   * @returns {void}
+   */
+  set(key, value = null) {
+    if (!this.settingConfigs[key])
+      throw new Error(`Invalid runtime setting: ${key}`);
+    this.settings[key] = this.settingConfigs[key].validate(value);
+  }
+}
+
+module.exports = RuntimeSettings;

collector/utils/url/index.js

@@ -1,3 +1,4 @@
+const RuntimeSettings = require("../runtimeSettings");
 /**  ATTN: SECURITY RESEARCHERS
  * To Security researchers about to submit an SSRF report CVE - please don't.
  * We are aware that the code below is does not defend against any of the thousands of ways
@@ -13,15 +14,24 @@
 
 const VALID_PROTOCOLS = ["https:", "http:"];
 const INVALID_OCTETS = [192, 172, 10, 127];
+const runtimeSettings = new RuntimeSettings();
 
 /**
  * If an ip address is passed in the user is attempting to collector some internal service running on internal/private IP.
  * This is not a security feature and simply just prevents the user from accidentally entering invalid IP addresses.
+ * Can be bypassed via COLLECTOR_ALLOW_ANY_IP environment variable.
  * @param {URL} param0
  * @param {URL['hostname']} param0.hostname
  * @returns {boolean}
  */
 function isInvalidIp({ hostname }) {
+  if (runtimeSettings.get("allowAnyIp")) {
+    console.log(
+      "\x1b[33mURL IP local address restrictions have been disabled by administrator!\x1b[0m"
+    );
+    return false;
+  }
+
   const IPRegex = new RegExp(
     /^(([0-9]|[1-9][0-9]|1[0-9]{2}|2[0-4][0-9]|25[0-5])\.){3}([0-9]|[1-9][0-9]|1[0-9]{2}|2[0-4][0-9]|25[0-5])$/gi
   );
@@ -40,6 +50,14 @@ function isInvalidIp({ hostname }) {
   return INVALID_OCTETS.includes(Number(octetOne));
 }
 
+/**
+ * Validates a URL
+ * - Checks the URL forms a valid URL
+ * - Checks the URL is at least HTTP(S)
+ * - Checks the URL is not an internal IP - can be bypassed via COLLECTOR_ALLOW_ANY_IP
+ * @param {string} url
+ * @returns {boolean}
+ */
 function validURL(url) {
   try {
     const destination = new URL(url);

docker/.env.example

@@ -322,6 +322,10 @@ GID='1000'
 # See https://docs.anythingllm.com/configuration#simple-sso-passthrough for more information.
 # SIMPLE_SSO_ENABLED=1
 
+# Allow scraping of any IP address in collector - must be string "true" to be enabled
+# See https://docs.anythingllm.com/configuration#local-ip-address-scraping for more information.
+# COLLECTOR_ALLOW_ANY_IP="true"
+
 # Specify the target languages for when using OCR to parse images and PDFs.
 # This is a comma separated list of language codes as a string. Unsupported languages will be ignored.
 # Default is English. See https://tesseract-ocr.github.io/tessdoc/Data-Files-in-different-versions.html for a list of valid language codes.

server/.env.example

@@ -311,6 +311,10 @@ TTS_PROVIDER="native"
 # See https://docs.anythingllm.com/configuration#simple-sso-passthrough for more information.
 # SIMPLE_SSO_ENABLED=1
 
+# Allow scraping of any IP address in collector - must be string "true" to be enabled
+# See https://docs.anythingllm.com/configuration#local-ip-address-scraping for more information.
+# COLLECTOR_ALLOW_ANY_IP="true"
+
 # Specify the target languages for when using OCR to parse images and PDFs.
 # This is a comma separated list of language codes as a string. Unsupported languages will be ignored.
 # Default is English. See https://tesseract-ocr.github.io/tessdoc/Data-Files-in-different-versions.html for a list of valid language codes.

server/utils/collectorApi/index.js

@@ -1,5 +1,14 @@
 const { EncryptionManager } = require("../EncryptionManager");
 
+/**
+ * @typedef {Object} CollectorOptions
+ * @property {string} whisperProvider - The provider to use for whisper, defaults to "local"
+ * @property {string} WhisperModelPref - The model to use for whisper if set.
+ * @property {string} openAiKey - The API key to use for OpenAI interfacing, mostly passed to OAI Whisper provider.
+ * @property {Object} ocr - The OCR options
+ * @property {{allowAnyIp: "true"|null|undefined}} runtimeSettings - The runtime settings that are passed to the collector. Persisted across requests.
+ */
+
 // When running locally will occupy the 0.0.0.0 hostname space but when deployed inside
 // of docker this endpoint is not exposed so it is only on the Docker instances internal network
 // so no additional security is needed on the endpoint directly. Auth is done however by the express
@@ -15,6 +24,10 @@ class CollectorApi {
     console.log(`\x1b[36m[CollectorApi]\x1b[0m ${text}`, ...args);
   }
 
+  /**
+   * Attach options to the request passed to the collector API
+   * @returns {CollectorOptions}
+   */
   #attachOptions() {
     return {
       whisperProvider: process.env.WHISPER_PROVIDER || "local",
@@ -23,6 +36,9 @@ class CollectorApi {
       ocr: {
         langList: process.env.TARGET_OCR_LANG || "eng",
       },
+      runtimeSettings: {
+        allowAnyIp: process.env.COLLECTOR_ALLOW_ANY_IP ?? "false",
+      },
     };
   }
 
@@ -45,6 +61,12 @@ class CollectorApi {
       });
   }
 
+  /**
+   * Process a document
+   * - Will append the options to the request body
+   * @param {string} filename - The filename of the document to process
+   * @returns {Promise<Object>} - The response from the collector API
+   */
   async processDocument(filename = "") {
     if (!filename) return false;
 
@@ -75,10 +97,16 @@ class CollectorApi {
       });
   }
 
+  /**
+   * Process a link
+   * - Will append the options to the request body
+   * @param {string} link - The link to process
+   * @returns {Promise<Object>} - The response from the collector API
+   */
   async processLink(link = "") {
     if (!link) return false;
 
-    const data = JSON.stringify({ link });
+    const data = JSON.stringify({ link, options: this.#attachOptions() });
     return await fetch(`${this.endpoint}/process-link`, {
       method: "POST",
       headers: {
@@ -101,8 +129,19 @@ class CollectorApi {
       });
   }
 
+  /**
+   * Process raw text as a document for the collector
+   * - Will append the options to the request body
+   * @param {string} textContent - The text to process
+   * @param {Object} metadata - The metadata to process
+   * @returns {Promise<Object>} - The response from the collector API
+   */
   async processRawText(textContent = "", metadata = {}) {
-    const data = JSON.stringify({ textContent, metadata });
+    const data = JSON.stringify({
+      textContent,
+      metadata,
+      options: this.#attachOptions(),
+    });
     return await fetch(`${this.endpoint}/process-raw-text`, {
       method: "POST",
       headers: {
@@ -151,10 +190,21 @@ class CollectorApi {
       });
   }
 
+  /**
+   * Get the content of a link only in a specific format
+   * - Will append the options to the request body
+   * @param {string} link - The link to get the content of
+   * @param {"text"|"html"} captureAs - The format to capture the content as
+   * @returns {Promise<Object>} - The response from the collector API
+   */
   async getLinkContent(link = "", captureAs = "text") {
     if (!link) return false;
 
-    const data = JSON.stringify({ link, captureAs });
+    const data = JSON.stringify({
+      link,
+      captureAs,
+      options: this.#attachOptions(),
+    });
     return await fetch(`${this.endpoint}/util/get-link`, {
       method: "POST",
       headers: {

server/utils/helpers/updateENV.js

@@ -958,6 +958,9 @@ function dumpENV() {
 
     // OCR Language Support
     "TARGET_OCR_LANG",
+
+    // Collector API common ENV - allows bypassing URL validation checks
+    "COLLECTOR_ALLOW_ANY_IP",
   ];
 
   // Simple sanitization of each value to prevent ENV injection via newline or quote escaping.