fix(knowledge): surface KB description validation errors and raise limit to 10k (#5347)

Author: TheodoreSpeaks

apps/sim/app/workspace/[workspaceId]/knowledge/components/create-base-modal/create-base-modal.tsx

@@ -17,14 +17,16 @@ import {
   type ComboboxOption,
   cn,
   Loader,
+  toast,
 } from '@sim/emcn'
 import { createLogger } from '@sim/logger'
 import { getErrorMessage } from '@sim/utils/errors'
 import { X } from 'lucide-react'
 import { useParams } from 'next/navigation'
-import { useForm } from 'react-hook-form'
+import { type FieldErrors, useForm } from 'react-hook-form'
 import { z } from 'zod'
 import type { StrategyOptions } from '@/lib/chunkers/types'
+import { KNOWLEDGE_BASE_DESCRIPTION_MAX_LENGTH } from '@/lib/knowledge/constants'
 import { formatFileSize, validateKnowledgeBaseFile } from '@/lib/uploads/utils/file-utils'
 import { ACCEPT_ATTRIBUTE } from '@/lib/uploads/utils/validation'
 import { useKnowledgeUpload } from '@/app/workspace/[workspaceId]/knowledge/hooks/use-knowledge-upload'
@@ -58,7 +60,13 @@ const FormSchema = z
       .min(1, 'Name is required')
       .max(100, 'Name must be less than 100 characters')
       .refine((value) => value.trim().length > 0, 'Name cannot be empty'),
-    description: z.string().max(500, 'Description must be less than 500 characters').optional(),
+    description: z
+      .string()
+      .max(
+        KNOWLEDGE_BASE_DESCRIPTION_MAX_LENGTH,
+        `Description must be ${KNOWLEDGE_BASE_DESCRIPTION_MAX_LENGTH} characters or less`
+      )
+      .optional(),
     minChunkSize: z
       .number()
       .min(1, 'Min chunk size must be at least 1 character')
@@ -223,6 +231,15 @@ export const CreateBaseModal = memo(function CreateBaseModal({
   const isSubmitting =
     createKnowledgeBaseMutation.isPending || deleteKnowledgeBaseMutation.isPending || isUploading
 
+  const onInvalid = (formErrors: FieldErrors<FormInputValues>) => {
+    const firstMessage = Object.values(formErrors).find(
+      (fieldError) => typeof fieldError?.message === 'string'
+    )?.message
+    toast.error(
+      typeof firstMessage === 'string' ? firstMessage : 'Please fix the highlighted fields'
+    )
+  }
+
   const onSubmit = async (data: FormValues) => {
     setSubmitStatus(null)
 
@@ -292,7 +309,7 @@ export const CreateBaseModal = memo(function CreateBaseModal({
     <ChipModal open={open} onOpenChange={handleClose} srTitle='Create Knowledge Base' size='lg'>
       <ChipModalHeader onClose={() => handleClose(false)}>Create Knowledge Base</ChipModalHeader>
 
-      <form onSubmit={handleSubmit(onSubmit)} className='flex min-h-0 flex-1 flex-col'>
+      <form onSubmit={handleSubmit(onSubmit, onInvalid)} className='flex min-h-0 flex-1 flex-col'>
         <button type='submit' hidden disabled={isSubmitting || !nameValue?.trim()} />
         <ChipModalBody>
           <input
@@ -317,7 +334,7 @@ export const CreateBaseModal = memo(function CreateBaseModal({
             />
           </ChipModalField>
 
-          <ChipModalField type='custom' title='Description'>
+          <ChipModalField type='custom' title='Description' error={errors.description?.message}>
             <ChipTextarea
               placeholder='Describe this knowledge base (optional)'
               rows={4}
@@ -520,7 +537,7 @@ export const CreateBaseModal = memo(function CreateBaseModal({
                     : 'Creating...'
                 : 'Creating...'
               : 'Create',
-            onClick: handleSubmit(onSubmit),
+            onClick: handleSubmit(onSubmit, onInvalid),
             disabled: isSubmitting || !nameValue?.trim(),
           }}
         />

apps/sim/app/workspace/[workspaceId]/knowledge/components/edit-knowledge-base-modal/edit-knowledge-base-modal.tsx

@@ -8,9 +8,11 @@ import {
   ChipModalField,
   ChipModalFooter,
   ChipModalHeader,
+  toast,
 } from '@sim/emcn'
 import { createLogger } from '@sim/logger'
 import { getErrorMessage } from '@sim/utils/errors'
+import { KNOWLEDGE_BASE_DESCRIPTION_MAX_LENGTH } from '@/lib/knowledge/constants'
 import type { ChunkingConfig } from '@/lib/knowledge/types'
 
 const logger = createLogger('EditKnowledgeBaseModal')
@@ -60,31 +62,36 @@ export const EditKnowledgeBaseModal = memo(function EditKnowledgeBaseModal({
     }
   }
 
-  const validate = (): boolean => {
-    let valid = true
+  const validate = (): string | null => {
+    let firstError: string | null = null
 
     if (!name.trim()) {
       setNameError('Name is required')
-      valid = false
+      firstError ??= 'Name is required'
     } else if (name.trim().length > 100) {
       setNameError('Name must be less than 100 characters')
-      valid = false
+      firstError ??= 'Name must be less than 100 characters'
     } else {
       setNameError(null)
     }
 
-    if (description.length > 500) {
-      setDescriptionError('Description must be less than 500 characters')
-      valid = false
+    if (description.length > KNOWLEDGE_BASE_DESCRIPTION_MAX_LENGTH) {
+      const message = `Description must be ${KNOWLEDGE_BASE_DESCRIPTION_MAX_LENGTH} characters or less`
+      setDescriptionError(message)
+      firstError ??= message
     } else {
       setDescriptionError(null)
     }
 
-    return valid
+    return firstError
   }
 
   const handleSubmit = async () => {
-    if (!validate()) return
+    const validationError = validate()
+    if (validationError) {
+      toast.error(validationError)
+      return
+    }
 
     setIsSubmitting(true)
     setError(null)

apps/sim/lib/api/contracts/knowledge/base.ts

@@ -7,6 +7,7 @@ import {
 } from '@/lib/api/contracts/knowledge/shared'
 import { defineRouteContract } from '@/lib/api/contracts/types'
 import type { StrategyOptions } from '@/lib/chunkers/types'
+import { KNOWLEDGE_BASE_DESCRIPTION_MAX_LENGTH } from '@/lib/knowledge/constants'
 
 export const knowledgeScopeSchema = z.enum(['active', 'archived', 'all'])
 export type KnowledgeScope = z.output<typeof knowledgeScopeSchema>
@@ -51,7 +52,13 @@ export const chunkingConfigSchema = z
 
 export const createKnowledgeBaseBodySchema = z.object({
   name: z.string().min(1, 'Name is required'),
-  description: z.string().optional(),
+  description: z
+    .string()
+    .max(
+      KNOWLEDGE_BASE_DESCRIPTION_MAX_LENGTH,
+      `Description must be ${KNOWLEDGE_BASE_DESCRIPTION_MAX_LENGTH} characters or less`
+    )
+    .optional(),
   workspaceId: z.string().min(1, 'Workspace ID is required'),
   embeddingModel: z.literal('text-embedding-3-small').default('text-embedding-3-small'),
   embeddingDimension: z.literal(1536).default(1536),

apps/sim/lib/api/contracts/v1/knowledge/index.ts

@@ -5,6 +5,7 @@ import {
   successResponseSchema,
 } from '@/lib/api/contracts/knowledge/shared'
 import { defineRouteContract } from '@/lib/api/contracts/types'
+import { KNOWLEDGE_BASE_DESCRIPTION_MAX_LENGTH } from '@/lib/knowledge/constants'
 
 /**
  * Public API v1 schemas (`/api/v1/knowledge/**`)
@@ -36,7 +37,13 @@ export const v1ListKnowledgeBasesQuerySchema = z.object({
 export const v1CreateKnowledgeBaseBodySchema = z.object({
   workspaceId: z.string().min(1, 'Workspace ID is required'),
   name: z.string().min(1, 'Name is required').max(255, 'Name must be 255 characters or less'),
-  description: z.string().max(1000, 'Description must be 1000 characters or less').optional(),
+  description: z
+    .string()
+    .max(
+      KNOWLEDGE_BASE_DESCRIPTION_MAX_LENGTH,
+      `Description must be ${KNOWLEDGE_BASE_DESCRIPTION_MAX_LENGTH} characters or less`
+    )
+    .optional(),
   chunkingConfig: v1ChunkingConfigSchema.optional().default({
     maxSize: 1024,
     minSize: 100,
@@ -54,7 +61,13 @@ export const v1UpdateKnowledgeBaseBodySchema = z
   .object({
     workspaceId: z.string().min(1, 'Workspace ID is required'),
     name: z.string().min(1).max(255, 'Name must be 255 characters or less').optional(),
-    description: z.string().max(1000, 'Description must be 1000 characters or less').optional(),
+    description: z
+      .string()
+      .max(
+        KNOWLEDGE_BASE_DESCRIPTION_MAX_LENGTH,
+        `Description must be ${KNOWLEDGE_BASE_DESCRIPTION_MAX_LENGTH} characters or less`
+      )
+      .optional(),
     chunkingConfig: z
       .object({
         maxSize: z.number().min(100).max(4000),

apps/sim/lib/knowledge/constants.ts

@@ -1,3 +1,6 @@
+/** Max character length for a knowledge base description, enforced at every layer (UI, internal API, v1 API). */
+export const KNOWLEDGE_BASE_DESCRIPTION_MAX_LENGTH = 10_000
+
 export const TAG_SLOT_CONFIG = {
   text: {
     slots: ['tag1', 'tag2', 'tag3', 'tag4', 'tag5', 'tag6', 'tag7'] as const,