improvement(security): added input validation for airtable, lemlist, and more tools to protect against SSRF (#2847)

waleedlatif1 · web-flow · commit 55700b9bf4ff · 2026-01-15T19:33:29.000-08:00
diff --git a/apps/sim/lib/core/security/input-validation.test.ts b/apps/sim/lib/core/security/input-validation.test.ts
@@ -2,6 +2,7 @@ import { loggerMock } from '@sim/testing'
 import { describe, expect, it, vi } from 'vitest'
 import {
   createPinnedUrl,
+  validateAirtableId,
   validateAlphanumericId,
   validateEnum,
   validateExternalUrl,
@@ -1112,3 +1113,82 @@ describe('validateGoogleCalendarId', () => {
     })
   })
 })
+
+describe('validateAirtableId', () => {
+  describe('valid base IDs (app prefix)', () => {
+    it.concurrent('should accept valid base ID', () => {
+      const result = validateAirtableId('appABCDEFGHIJKLMN', 'app', 'baseId')
+      expect(result.isValid).toBe(true)
+      expect(result.sanitized).toBe('appABCDEFGHIJKLMN')
+    })
+
+    it.concurrent('should accept base ID with mixed case', () => {
+      const result = validateAirtableId('appAbCdEfGhIjKlMn', 'app', 'baseId')
+      expect(result.isValid).toBe(true)
+    })
+
+    it.concurrent('should accept base ID with numbers', () => {
+      const result = validateAirtableId('app12345678901234', 'app', 'baseId')
+      expect(result.isValid).toBe(true)
+    })
+  })
+
+  describe('valid table IDs (tbl prefix)', () => {
+    it.concurrent('should accept valid table ID', () => {
+      const result = validateAirtableId('tblABCDEFGHIJKLMN', 'tbl', 'tableId')
+      expect(result.isValid).toBe(true)
+    })
+  })
+
+  describe('valid webhook IDs (ach prefix)', () => {
+    it.concurrent('should accept valid webhook ID', () => {
+      const result = validateAirtableId('achABCDEFGHIJKLMN', 'ach', 'webhookId')
+      expect(result.isValid).toBe(true)
+    })
+  })
+
+  describe('invalid IDs', () => {
+    it.concurrent('should reject null', () => {
+      const result = validateAirtableId(null, 'app', 'baseId')
+      expect(result.isValid).toBe(false)
+      expect(result.error).toContain('required')
+    })
+
+    it.concurrent('should reject empty string', () => {
+      const result = validateAirtableId('', 'app', 'baseId')
+      expect(result.isValid).toBe(false)
+      expect(result.error).toContain('required')
+    })
+
+    it.concurrent('should reject wrong prefix', () => {
+      const result = validateAirtableId('tblABCDEFGHIJKLMN', 'app', 'baseId')
+      expect(result.isValid).toBe(false)
+      expect(result.error).toContain('starting with "app"')
+    })
+
+    it.concurrent('should reject too short ID (13 chars after prefix)', () => {
+      const result = validateAirtableId('appABCDEFGHIJKLM', 'app', 'baseId')
+      expect(result.isValid).toBe(false)
+    })
+
+    it.concurrent('should reject too long ID (15 chars after prefix)', () => {
+      const result = validateAirtableId('appABCDEFGHIJKLMNO', 'app', 'baseId')
+      expect(result.isValid).toBe(false)
+    })
+
+    it.concurrent('should reject special characters', () => {
+      const result = validateAirtableId('appABCDEFGH/JKLMN', 'app', 'baseId')
+      expect(result.isValid).toBe(false)
+    })
+
+    it.concurrent('should reject path traversal attempts', () => {
+      const result = validateAirtableId('app../etc/passwd', 'app', 'baseId')
+      expect(result.isValid).toBe(false)
+    })
+
+    it.concurrent('should reject lowercase prefix', () => {
+      const result = validateAirtableId('AppABCDEFGHIJKLMN', 'app', 'baseId')
+      expect(result.isValid).toBe(false)
+    })
+  })
+})
diff --git a/apps/sim/lib/core/security/input-validation.ts b/apps/sim/lib/core/security/input-validation.ts
@@ -896,6 +896,57 @@ export function createPinnedUrl(originalUrl: string, resolvedIP: string): string
   return `${parsed.protocol}//${resolvedIP}${port}${parsed.pathname}${parsed.search}`
 }
 
+/**
+ * Validates an Airtable ID (base, table, or webhook ID)
+ *
+ * Airtable IDs have specific prefixes:
+ * - Base IDs: "app" + 14 alphanumeric characters (e.g., appXXXXXXXXXXXXXX)
+ * - Table IDs: "tbl" + 14 alphanumeric characters
+ * - Webhook IDs: "ach" + 14 alphanumeric characters
+ *
+ * @param value - The ID to validate
+ * @param expectedPrefix - The expected prefix ('app', 'tbl', or 'ach')
+ * @param paramName - Name of the parameter for error messages
+ * @returns ValidationResult
+ *
+ * @example
+ * ```typescript
+ * const result = validateAirtableId(baseId, 'app', 'baseId')
+ * if (!result.isValid) {
+ *   throw new Error(result.error)
+ * }
+ * ```
+ */
+export function validateAirtableId(
+  value: string | null | undefined,
+  expectedPrefix: 'app' | 'tbl' | 'ach',
+  paramName = 'ID'
+): ValidationResult {
+  if (value === null || value === undefined || value === '') {
+    return {
+      isValid: false,
+      error: `${paramName} is required`,
+    }
+  }
+
+  // Airtable IDs: prefix (3 chars) + 14 alphanumeric characters = 17 chars total
+  const airtableIdPattern = new RegExp(`^${expectedPrefix}[a-zA-Z0-9]{14}$`)
+
+  if (!airtableIdPattern.test(value)) {
+    logger.warn('Invalid Airtable ID format', {
+      paramName,
+      expectedPrefix,
+      value: value.substring(0, 20),
+    })
+    return {
+      isValid: false,
+      error: `${paramName} must be a valid Airtable ID starting with "${expectedPrefix}"`,
+    }
+  }
+
+  return { isValid: true, sanitized: value }
+}
+
 /**
  * Validates a Google Calendar ID
  *
diff --git a/apps/sim/lib/webhooks/provider-subscriptions.ts b/apps/sim/lib/webhooks/provider-subscriptions.ts
@@ -1,5 +1,6 @@
 import { createLogger } from '@sim/logger'
 import type { NextRequest } from 'next/server'
+import { validateAirtableId, validateAlphanumericId } from '@/lib/core/security/input-validation'
 import { getBaseUrl } from '@/lib/core/utils/urls'
 import { getOAuthToken, refreshAccessTokenIfNeeded } from '@/app/api/auth/oauth/utils'
 
@@ -358,6 +359,15 @@ export async function deleteAirtableWebhook(
       return
     }
 
+    const baseIdValidation = validateAirtableId(baseId, 'app', 'baseId')
+    if (!baseIdValidation.isValid) {
+      airtableLogger.warn(`[${requestId}] Invalid Airtable base ID format, skipping deletion`, {
+        webhookId: webhook.id,
+        baseId: baseId.substring(0, 20),
+      })
+      return
+    }
+
     const userIdForToken = workflow.userId
     const accessToken = await getOAuthToken(userIdForToken, 'airtable')
     if (!accessToken) {
@@ -428,6 +438,15 @@ export async function deleteAirtableWebhook(
       return
     }
 
+    const webhookIdValidation = validateAirtableId(resolvedExternalId, 'ach', 'webhookId')
+    if (!webhookIdValidation.isValid) {
+      airtableLogger.warn(`[${requestId}] Invalid Airtable webhook ID format, skipping deletion`, {
+        webhookId: webhook.id,
+        externalId: resolvedExternalId.substring(0, 20),
+      })
+      return
+    }
+
     const airtableDeleteUrl = `https://api.airtable.com/v0/bases/${baseId}/webhooks/${resolvedExternalId}`
     const airtableResponse = await fetch(airtableDeleteUrl, {
       method: 'DELETE',
@@ -732,6 +751,14 @@ export async function deleteLemlistWebhook(webhook: any, requestId: string): Pro
     const authString = Buffer.from(`:${apiKey}`).toString('base64')
 
     const deleteById = async (id: string) => {
+      const validation = validateAlphanumericId(id, 'Lemlist hook ID', 50)
+      if (!validation.isValid) {
+        lemlistLogger.warn(`[${requestId}] Invalid Lemlist hook ID format, skipping deletion`, {
+          id: id.substring(0, 30),
+        })
+        return
+      }
+
       const lemlistApiUrl = `https://api.lemlist.com/api/hooks/${id}`
       const lemlistResponse = await fetch(lemlistApiUrl, {
         method: 'DELETE',
@@ -823,6 +850,24 @@ export async function deleteWebflowWebhook(
       return
     }
 
+    const siteIdValidation = validateAlphanumericId(siteId, 'siteId', 100)
+    if (!siteIdValidation.isValid) {
+      webflowLogger.warn(`[${requestId}] Invalid Webflow site ID format, skipping deletion`, {
+        webhookId: webhook.id,
+        siteId: siteId.substring(0, 30),
+      })
+      return
+    }
+
+    const webhookIdValidation = validateAlphanumericId(externalId, 'webhookId', 100)
+    if (!webhookIdValidation.isValid) {
+      webflowLogger.warn(`[${requestId}] Invalid Webflow webhook ID format, skipping deletion`, {
+        webhookId: webhook.id,
+        externalId: externalId.substring(0, 30),
+      })
+      return
+    }
+
     const accessToken = await getOAuthToken(workflow.userId, 'webflow')
     if (!accessToken) {
       webflowLogger.warn(
@@ -1122,6 +1167,16 @@ export async function createAirtableWebhookSubscription(
       )
     }
 
+    const baseIdValidation = validateAirtableId(baseId, 'app', 'baseId')
+    if (!baseIdValidation.isValid) {
+      throw new Error(baseIdValidation.error)
+    }
+
+    const tableIdValidation = validateAirtableId(tableId, 'tbl', 'tableId')
+    if (!tableIdValidation.isValid) {
+      throw new Error(tableIdValidation.error)
+    }
+
     const accessToken = await getOAuthToken(userId, 'airtable')
     if (!accessToken) {
       airtableLogger.warn(
@@ -1354,6 +1409,11 @@ export async function createWebflowWebhookSubscription(
       throw new Error('Site ID is required to create Webflow webhook')
     }
 
+    const siteIdValidation = validateAlphanumericId(siteId, 'siteId', 100)
+    if (!siteIdValidation.isValid) {
+      throw new Error(siteIdValidation.error)
+    }
+
     if (!triggerId) {
       webflowLogger.warn(`[${requestId}] Missing triggerId for Webflow webhook creation.`, {
         webhookId: webhookData.id,
diff --git a/apps/sim/tools/pulse/parser.ts b/apps/sim/tools/pulse/parser.ts
@@ -86,7 +86,6 @@ export const pulseParserTool: ToolConfig<PulseParserInput, PulseParserOutput> =
         throw new Error('Missing or invalid API key: A valid Pulse API key is required')
       }
 
-      // Check if we have a file upload instead of direct URL
       if (
         params.fileUpload &&
         (!params.filePath || params.filePath === 'null' || params.filePath === '')
@@ -137,13 +136,6 @@ export const pulseParserTool: ToolConfig<PulseParserInput, PulseParserOutput> =
         if (!['http:', 'https:'].includes(url.protocol)) {
           throw new Error(`Invalid protocol: ${url.protocol}. URL must use HTTP or HTTPS protocol`)
         }
-
-        if (url.hostname.includes('drive.google.com') || url.hostname.includes('docs.google.com')) {
-          throw new Error(
-            'Google Drive links are not supported. ' +
-              'Please upload your document or provide a direct download link.'
-          )
-        }
       } catch (error) {
         const errorMessage = error instanceof Error ? error.message : String(error)
         throw new Error(
@@ -156,12 +148,10 @@ export const pulseParserTool: ToolConfig<PulseParserInput, PulseParserOutput> =
         filePath: url.toString(),
       }
 
-      // Check if this is an internal workspace file path
       if (params.fileUpload?.path?.startsWith('/api/files/serve/')) {
         requestBody.filePath = params.fileUpload.path
       }
 
-      // Add optional parameters
       if (params.pages && typeof params.pages === 'string' && params.pages.trim() !== '') {
         requestBody.pages = params.pages.trim()
       }
@@ -204,7 +194,6 @@ export const pulseParserTool: ToolConfig<PulseParserInput, PulseParserOutput> =
       throw new Error('Invalid response format from Pulse API')
     }
 
-    // Pass through the native Pulse API response
     const pulseData =
       parseResult.output && typeof parseResult.output === 'object'
         ? parseResult.output
diff --git a/apps/sim/tools/reducto/parser.ts b/apps/sim/tools/reducto/parser.ts
@@ -63,7 +63,6 @@ export const reductoParserTool: ToolConfig<ReductoParserInput, ReductoParserOutp
         throw new Error('Missing or invalid API key: A valid Reducto API key is required')
       }
 
-      // Check if we have a file upload instead of direct URL
       if (
         params.fileUpload &&
         (!params.filePath || params.filePath === 'null' || params.filePath === '')
@@ -110,13 +109,6 @@ export const reductoParserTool: ToolConfig<ReductoParserInput, ReductoParserOutp
         if (!['http:', 'https:'].includes(url.protocol)) {
           throw new Error(`Invalid protocol: ${url.protocol}. URL must use HTTP or HTTPS protocol`)
         }
-
-        if (url.hostname.includes('drive.google.com') || url.hostname.includes('docs.google.com')) {
-          throw new Error(
-            'Google Drive links are not supported by the Reducto API. ' +
-              'Please upload your PDF to a public web server or provide a direct download link.'
-          )
-        }
       } catch (error) {
         const errorMessage = error instanceof Error ? error.message : String(error)
         throw new Error(
@@ -129,7 +121,6 @@ export const reductoParserTool: ToolConfig<ReductoParserInput, ReductoParserOutp
         filePath: url.toString(),
       }
 
-      // Check if this is an internal workspace file path
       if (params.fileUpload?.path?.startsWith('/api/files/serve/')) {
         requestBody.filePath = params.fileUpload.path
       }
@@ -138,7 +129,6 @@ export const reductoParserTool: ToolConfig<ReductoParserInput, ReductoParserOutp
         requestBody.tableOutputFormat = params.tableOutputFormat
       }
 
-      // Page selection
       if (params.pages !== undefined && params.pages !== null) {
         if (Array.isArray(params.pages) && params.pages.length > 0) {
           const validPages = params.pages.filter(
@@ -162,7 +152,6 @@ export const reductoParserTool: ToolConfig<ReductoParserInput, ReductoParserOutp
       throw new Error('Invalid response format from Reducto API')
     }
 
-    // Pass through the native Reducto response
     const reductoData = data.output ?? data
 
     return {
