Merge pull request #590 from lfnovo/feat/error-clarity-handling

feat: improve error clarity for LLM provider failures

CHANGELOG.md

@@ -7,6 +7,22 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [1.7.2] - 2026-02-16
+
+### Added
+- Error classification utility that maps LLM provider errors to user-friendly messages (#506)
+- Global exception handlers in FastAPI for all custom exception types with proper HTTP status codes
+- `getApiErrorMessage()` frontend helper that falls back to backend messages when no i18n mapping exists
+
+### Fixed
+- LLM errors (invalid API key, wrong model, rate limits) now show descriptive messages instead of "An unexpected error occurred"
+- SSE streaming error events in source chat and ask hooks were swallowed by inner JSON parse catch blocks
+- Transformation execution errors were caught and re-wrapped as generic 500s instead of using proper status codes
+
+### Changed
+- `ValueError` replaced with `ConfigurationError` in model provisioning for proper error classification
+- `ConfigurationError` added to command retry `stop_on` lists to avoid retrying permanent config failures
+
 ## [1.7.1] - 2026-02-14
 
 ### Added

CLAUDE.md

@@ -218,4 +218,4 @@ See dedicated CLAUDE.md files for detailed guidance:
 
 ---
 
-**Last Updated**: February 2026 | **Project Version**: 1.7.1
+**Last Updated**: February 2026 | **Project Version**: 1.7.2

api/main.py

@@ -12,6 +12,16 @@ from loguru import logger
 from starlette.exceptions import HTTPException as StarletteHTTPException
 
 from api.auth import PasswordAuthMiddleware
+from open_notebook.exceptions import (
+    AuthenticationError,
+    ConfigurationError,
+    ExternalServiceError,
+    InvalidInputError,
+    NetworkError,
+    NotFoundError,
+    OpenNotebookError,
+    RateLimitError,
+)
 from api.routers import (
     auth,
     chat,
@@ -154,6 +164,88 @@ async def custom_http_exception_handler(request: Request, exc: StarletteHTTPExce
     )
 
 
+def _cors_headers(request: Request) -> dict[str, str]:
+    origin = request.headers.get("origin", "*")
+    return {
+        "Access-Control-Allow-Origin": origin,
+        "Access-Control-Allow-Credentials": "true",
+        "Access-Control-Allow-Methods": "*",
+        "Access-Control-Allow-Headers": "*",
+    }
+
+
+@app.exception_handler(NotFoundError)
+async def not_found_error_handler(request: Request, exc: NotFoundError):
+    return JSONResponse(
+        status_code=404,
+        content={"detail": str(exc)},
+        headers=_cors_headers(request),
+    )
+
+
+@app.exception_handler(InvalidInputError)
+async def invalid_input_error_handler(request: Request, exc: InvalidInputError):
+    return JSONResponse(
+        status_code=400,
+        content={"detail": str(exc)},
+        headers=_cors_headers(request),
+    )
+
+
+@app.exception_handler(AuthenticationError)
+async def authentication_error_handler(request: Request, exc: AuthenticationError):
+    return JSONResponse(
+        status_code=401,
+        content={"detail": str(exc)},
+        headers=_cors_headers(request),
+    )
+
+
+@app.exception_handler(RateLimitError)
+async def rate_limit_error_handler(request: Request, exc: RateLimitError):
+    return JSONResponse(
+        status_code=429,
+        content={"detail": str(exc)},
+        headers=_cors_headers(request),
+    )
+
+
+@app.exception_handler(ConfigurationError)
+async def configuration_error_handler(request: Request, exc: ConfigurationError):
+    return JSONResponse(
+        status_code=422,
+        content={"detail": str(exc)},
+        headers=_cors_headers(request),
+    )
+
+
+@app.exception_handler(NetworkError)
+async def network_error_handler(request: Request, exc: NetworkError):
+    return JSONResponse(
+        status_code=502,
+        content={"detail": str(exc)},
+        headers=_cors_headers(request),
+    )
+
+
+@app.exception_handler(ExternalServiceError)
+async def external_service_error_handler(request: Request, exc: ExternalServiceError):
+    return JSONResponse(
+        status_code=502,
+        content={"detail": str(exc)},
+        headers=_cors_headers(request),
+    )
+
+
+@app.exception_handler(OpenNotebookError)
+async def open_notebook_error_handler(request: Request, exc: OpenNotebookError):
+    return JSONResponse(
+        status_code=500,
+        content={"detail": str(exc)},
+        headers=_cors_headers(request),
+    )
+
+
 # Include routers
 app.include_router(auth.router, prefix="/api", tags=["auth"])
 app.include_router(config.router, prefix="/api", tags=["config"])

api/routers/search.py

@@ -102,8 +102,11 @@ async def stream_ask_response(
         yield f"data: {json.dumps(completion_data)}\n\n"
 
     except Exception as e:
+        from open_notebook.utils.error_classifier import classify_error
+
+        _, user_message = classify_error(e)
         logger.error(f"Error in ask streaming: {str(e)}")
-        error_data = {"type": "error", "message": str(e)}
+        error_data = {"type": "error", "message": user_message}
         yield f"data: {json.dumps(error_data)}\n\n"
 
 

api/routers/source_chat.py

@@ -470,8 +470,11 @@ async def stream_source_chat_response(
         yield f"data: {json.dumps(completion_event)}\n\n"
 
     except Exception as e:
+        from open_notebook.utils.error_classifier import classify_error
+
+        _, user_message = classify_error(e)
         logger.error(f"Error in source chat streaming: {str(e)}")
-        error_event = {"type": "error", "message": str(e)}
+        error_event = {"type": "error", "message": user_message}
         yield f"data: {json.dumps(error_event)}\n\n"
 
 

api/routers/transformations.py

@@ -14,7 +14,7 @@ from api.models import (
 )
 from open_notebook.ai.models import Model
 from open_notebook.domain.transformation import DefaultPrompts, Transformation
-from open_notebook.exceptions import InvalidInputError
+from open_notebook.exceptions import InvalidInputError, OpenNotebookError
 from open_notebook.graphs.transformation import graph as transformation_graph
 
 router = APIRouter()
@@ -109,6 +109,8 @@ async def execute_transformation(execute_request: TransformationExecuteRequest):
 
     except HTTPException:
         raise
+    except OpenNotebookError:
+        raise  # Let global exception handlers return proper status codes
     except Exception as e:
         logger.error(f"Error executing transformation: {str(e)}")
         raise HTTPException(

commands/embedding_commands.py

@@ -7,6 +7,7 @@ from surreal_commands import CommandInput, CommandOutput, command, submit_comman
 
 from open_notebook.ai.models import model_manager
 from open_notebook.database.repository import ensure_record_id, repo_insert, repo_query
+from open_notebook.exceptions import ConfigurationError
 from open_notebook.domain.notebook import Note, Source, SourceInsight
 from open_notebook.utils.chunking import ContentType, chunk_text, detect_content_type
 from open_notebook.utils.embedding import generate_embedding, generate_embeddings
@@ -125,7 +126,7 @@ class EmbedSourceOutput(CommandOutput):
         "wait_strategy": "exponential_jitter",
         "wait_min": 1,
         "wait_max": 60,
-        "stop_on": [ValueError],  # Don't retry validation errors
+        "stop_on": [ValueError, ConfigurationError],  # Don't retry validation/config errors
         "retry_log_level": "debug",
     },
 )
@@ -217,7 +218,7 @@ async def embed_note_command(input_data: EmbedNoteInput) -> EmbedNoteOutput:
         "wait_strategy": "exponential_jitter",
         "wait_min": 1,
         "wait_max": 60,
-        "stop_on": [ValueError],  # Don't retry validation errors
+        "stop_on": [ValueError, ConfigurationError],  # Don't retry validation/config errors
         "retry_log_level": "debug",
     },
 )
@@ -311,7 +312,7 @@ async def embed_insight_command(input_data: EmbedInsightInput) -> EmbedInsightOu
         "wait_strategy": "exponential_jitter",
         "wait_min": 1,
         "wait_max": 60,
-        "stop_on": [ValueError],  # Don't retry validation errors
+        "stop_on": [ValueError, ConfigurationError],  # Don't retry validation/config errors
         "retry_log_level": "debug",
     },
 )
@@ -447,7 +448,7 @@ async def embed_source_command(input_data: EmbedSourceInput) -> EmbedSourceOutpu
         "wait_strategy": "exponential_jitter",
         "wait_min": 1,
         "wait_max": 60,
-        "stop_on": [ValueError],  # Don't retry validation errors
+        "stop_on": [ValueError, ConfigurationError],  # Don't retry validation/config errors
         "retry_log_level": "debug",
     },
 )

commands/source_commands.py

@@ -8,6 +8,7 @@ from surreal_commands import CommandInput, CommandOutput, command
 from open_notebook.database.repository import ensure_record_id
 from open_notebook.domain.notebook import Source
 from open_notebook.domain.transformation import Transformation
+from open_notebook.exceptions import ConfigurationError
 
 try:
     from open_notebook.graphs.source import source_graph
@@ -53,7 +54,7 @@ class SourceProcessingOutput(CommandOutput):
         "wait_strategy": "exponential_jitter",
         "wait_min": 1,
         "wait_max": 120,  # Allow queue to drain
-        "stop_on": [ValueError],  # Don't retry validation errors
+        "stop_on": [ValueError, ConfigurationError],  # Don't retry validation/config errors
         "retry_log_level": "debug",  # Avoid log noise during transaction conflicts
     },
 )
@@ -184,7 +185,7 @@ class RunTransformationOutput(CommandOutput):
         "wait_strategy": "exponential_jitter",
         "wait_min": 1,
         "wait_max": 60,
-        "stop_on": [ValueError],  # Don't retry validation errors
+        "stop_on": [ValueError, ConfigurationError],  # Don't retry validation/config errors
         "retry_log_level": "debug",
     },
 )

frontend/src/lib/hooks/use-ask.ts

@@ -3,7 +3,7 @@
 import { useState, useCallback } from 'react'
 import { toast } from 'sonner'
 import { useTranslation } from '@/lib/hooks/use-translation'
-import { getApiErrorKey } from '@/lib/utils/error-handler'
+import { getApiErrorMessage } from '@/lib/utils/error-handler'
 import { searchApi } from '@/lib/api/search'
 import { AskStreamEvent } from '@/lib/types/search'
 
@@ -122,8 +122,12 @@ export function useAsk() {
                 throw new Error(data.message || 'Stream error occurred')
               }
             } catch (e) {
-              console.error('Error parsing SSE data:', e, 'Line:', line)
-              // Don't throw - continue processing other lines
+              if (e instanceof SyntaxError) {
+                console.error('Error parsing SSE data:', e, 'Line:', line)
+                // Don't throw - continue processing other lines
+              } else {
+                throw e
+              }
             }
           }
         }
@@ -144,7 +148,7 @@ export function useAsk() {
       }))
 
       toast.error(t('apiErrors.askFailed'), {
-        description: t(getApiErrorKey(errorMessage))
+        description: getApiErrorMessage(errorMessage, (key) => t(key))
       })
     }
   }, [t])

frontend/src/lib/hooks/use-sources.ts

@@ -4,7 +4,7 @@ import { sourcesApi } from '@/lib/api/sources'
 import { QUERY_KEYS } from '@/lib/api/query-client'
 import { useToast } from '@/lib/hooks/use-toast'
 import { useTranslation } from '@/lib/hooks/use-translation'
-import { getApiErrorKey } from '@/lib/utils/error-handler'
+import { getApiErrorMessage } from '@/lib/utils/error-handler'
 import {
   CreateSourceRequest,
   UpdateSourceRequest,
@@ -131,7 +131,7 @@ export function useCreateSource() {
     onError: (error: unknown) => {
       toast({
         title: t.common.error,
-        description: t(getApiErrorKey(error, t.sources.failedToAddSource)),
+        description: getApiErrorMessage(error, (key) => t(key), t.sources.failedToAddSource),
         variant: 'destructive',
       })
     },
@@ -158,7 +158,7 @@ export function useUpdateSource() {
     onError: (error: unknown) => {
       toast({
         title: t.common.error,
-        description: t(getApiErrorKey(error, t.sources.failedToUpdateSource)),
+        description: getApiErrorMessage(error, (key) => t(key), t.sources.failedToUpdateSource),
         variant: 'destructive',
       })
     },
@@ -185,7 +185,7 @@ export function useDeleteSource() {
     onError: (error: unknown) => {
       toast({
         title: t.common.error,
-        description: t(getApiErrorKey(error, t.sources.failedToDeleteSource)),
+        description: getApiErrorMessage(error, (key) => t(key), t.sources.failedToDeleteSource),
         variant: 'destructive',
       })
     },
@@ -212,7 +212,7 @@ export function useFileUpload() {
     onError: (error: unknown) => {
       toast({
         title: t.common.error,
-        description: t(getApiErrorKey(error, t.sources.failedToUploadFile)),
+        description: getApiErrorMessage(error, (key) => t(key), t.sources.failedToUploadFile),
         variant: 'destructive',
       })
     },
@@ -270,7 +270,7 @@ export function useRetrySource() {
     onError: (error: unknown) => {
       toast({
         title: t.common.error,
-        description: t(getApiErrorKey(error, t.sources.failedToRetry)),
+        description: getApiErrorMessage(error, (key) => t(key), t.sources.failedToRetry),
         variant: 'destructive',
       })
     },
@@ -332,7 +332,7 @@ export function useAddSourcesToNotebook() {
     onError: (error: unknown) => {
       toast({
         title: t.common.error,
-        description: t(getApiErrorKey(error, t.sources.failedToAddSourcesToNotebook)),
+        description: getApiErrorMessage(error, (key) => t(key), t.sources.failedToAddSourcesToNotebook),
         variant: 'destructive',
       })
     },
@@ -366,7 +366,7 @@ export function useRemoveSourceFromNotebook() {
     onError: (error: unknown) => {
       toast({
         title: t.common.error,
-        description: t(getApiErrorKey(error, t.sources.failedToRemoveSourceFromNotebook)),
+        description: getApiErrorMessage(error, (key) => t(key), t.sources.failedToRemoveSourceFromNotebook),
         variant: 'destructive',
       })
     },

frontend/src/lib/hooks/use-transformations.ts

@@ -2,7 +2,7 @@ import { useQuery, useMutation, useQueryClient } from '@tanstack/react-query'
 import { transformationsApi } from '@/lib/api/transformations'
 import { useToast } from '@/lib/hooks/use-toast'
 import { useTranslation } from '@/lib/hooks/use-translation'
-import { getApiErrorKey } from '@/lib/utils/error-handler'
+import { getApiErrorMessage } from '@/lib/utils/error-handler'
 import {
   CreateTransformationRequest,
   UpdateTransformationRequest,
@@ -49,7 +49,7 @@ export function useCreateTransformation() {
     onError: (error: unknown) => {
       toast({
         title: t.common.error,
-        description: t(getApiErrorKey(error, t.common.error)),
+        description: getApiErrorMessage(error, (key) => t(key)),
         variant: 'destructive',
       })
     },
@@ -75,7 +75,7 @@ export function useUpdateTransformation() {
     onError: (error: unknown) => {
       toast({
         title: t.common.error,
-        description: t(getApiErrorKey(error, t.common.error)),
+        description: getApiErrorMessage(error, (key) => t(key)),
         variant: 'destructive',
       })
     },
@@ -99,7 +99,7 @@ export function useDeleteTransformation() {
     onError: (error: unknown) => {
       toast({
         title: t.common.error,
-        description: t(getApiErrorKey(error, t.common.error)),
+        description: getApiErrorMessage(error, (key) => t(key)),
         variant: 'destructive',
       })
     },
@@ -115,7 +115,7 @@ export function useExecuteTransformation() {
     onError: (error: unknown) => {
       toast({
         title: t.common.error,
-        description: t(getApiErrorKey(error, t.common.error)),
+        description: getApiErrorMessage(error, (key) => t(key)),
         variant: 'destructive',
       })
     },
@@ -146,7 +146,7 @@ export function useUpdateDefaultPrompt() {
     onError: (error: unknown) => {
       toast({
         title: t.common.error,
-        description: t(getApiErrorKey(error, t.common.error)),
+        description: getApiErrorMessage(error, (key) => t(key)),
         variant: 'destructive',
       })
     },

frontend/src/lib/hooks/useNotebookChat.ts

@@ -3,7 +3,7 @@
 import { useState, useCallback, useEffect } from 'react'
 import { useQuery, useMutation, useQueryClient } from '@tanstack/react-query'
 import { toast } from 'sonner'
-import { getApiErrorKey } from '@/lib/utils/error-handler'
+import { getApiErrorMessage } from '@/lib/utils/error-handler'
 import { useTranslation } from '@/lib/hooks/use-translation'
 import { chatApi } from '@/lib/api/chat'
 import { QUERY_KEYS } from '@/lib/api/query-client'
@@ -84,7 +84,7 @@ export function useNotebookChat({ notebookId, sources, notes, contextSelections
     },
     onError: (err: unknown) => {
       const error = err as { response?: { data?: { detail?: string } }, message?: string };
-      toast.error(t(getApiErrorKey(error.response?.data?.detail || error.message, 'apiErrors.failedToCreateSession')))
+      toast.error(getApiErrorMessage(error.response?.data?.detail || error.message, (key) => t(key), 'apiErrors.failedToCreateSession'))
     }
   })
 
@@ -105,7 +105,7 @@ export function useNotebookChat({ notebookId, sources, notes, contextSelections
     },
     onError: (err: unknown) => {
       const error = err as { response?: { data?: { detail?: string } }, message?: string };
-      toast.error(t(getApiErrorKey(error.response?.data?.detail || error.message, 'apiErrors.failedToUpdateSession')))
+      toast.error(getApiErrorMessage(error.response?.data?.detail || error.message, (key) => t(key), 'apiErrors.failedToUpdateSession'))
     }
   })
 
@@ -125,7 +125,7 @@ export function useNotebookChat({ notebookId, sources, notes, contextSelections
     },
     onError: (err: unknown) => {
       const error = err as { response?: { data?: { detail?: string } }, message?: string };
-      toast.error(t(getApiErrorKey(error.response?.data?.detail || error.message, 'apiErrors.failedToDeleteSession')))
+      toast.error(getApiErrorMessage(error.response?.data?.detail || error.message, (key) => t(key), 'apiErrors.failedToDeleteSession'))
     }
   })
 
@@ -197,7 +197,7 @@ export function useNotebookChat({ notebookId, sources, notes, contextSelections
         })
       } catch (err: unknown) {
         const error = err as { response?: { data?: { detail?: string } }, message?: string };
-        toast.error(t(getApiErrorKey(error.response?.data?.detail || error.message, 'apiErrors.failedToCreateSession')))
+        toast.error(getApiErrorMessage(error.response?.data?.detail || error.message, (key) => t(key), 'apiErrors.failedToCreateSession'))
         return
       }
     }
@@ -230,7 +230,7 @@ export function useNotebookChat({ notebookId, sources, notes, contextSelections
     } catch (err: unknown) {
       const error = err as { response?: { data?: { detail?: string } }, message?: string };
       console.error('Error sending message:', error)
-      toast.error(t(getApiErrorKey(error.response?.data?.detail || error.message, 'apiErrors.failedToSendMessage')))
+      toast.error(getApiErrorMessage(error.response?.data?.detail || error.message, (key) => t(key), 'apiErrors.failedToSendMessage'))
       // Remove optimistic message on error
       setMessages(prev => prev.filter(msg => !msg.id.startsWith('temp-')))
     } finally {

frontend/src/lib/hooks/useSourceChat.ts

@@ -3,7 +3,7 @@
 import { useState, useCallback, useRef, useEffect } from 'react'
 import { useQuery, useMutation, useQueryClient } from '@tanstack/react-query'
 import { toast } from 'sonner'
-import { getApiErrorKey } from '@/lib/utils/error-handler'
+import { getApiErrorMessage } from '@/lib/utils/error-handler'
 import { useTranslation } from '@/lib/hooks/use-translation'
 import { sourceChatApi } from '@/lib/api/source-chat'
 import {
@@ -64,7 +64,7 @@ export function useSourceChat(sourceId: string) {
     },
     onError: (err: unknown) => {
       const error = err as { response?: { data?: { detail?: string } }, message?: string };
-      toast.error(t(getApiErrorKey(error.response?.data?.detail || error.message, 'apiErrors.failedToCreateSession')))
+      toast.error(getApiErrorMessage(error.response?.data?.detail || error.message, (key) => t(key), 'apiErrors.failedToCreateSession'))
     }
   })
 
@@ -79,7 +79,7 @@ export function useSourceChat(sourceId: string) {
     },
     onError: (err: unknown) => {
       const error = err as { response?: { data?: { detail?: string } }, message?: string };
-      toast.error(t(getApiErrorKey(error.response?.data?.detail || error.message, 'apiErrors.failedToUpdateSession')))
+      toast.error(getApiErrorMessage(error.response?.data?.detail || error.message, (key) => t(key), 'apiErrors.failedToUpdateSession'))
     }
   })
 
@@ -97,7 +97,7 @@ export function useSourceChat(sourceId: string) {
     },
     onError: (err: unknown) => {
       const error = err as { response?: { data?: { detail?: string } }, message?: string };
-      toast.error(t(getApiErrorKey(error.response?.data?.detail || error.message, 'apiErrors.failedToDeleteSession')))
+      toast.error(getApiErrorMessage(error.response?.data?.detail || error.message, (key) => t(key), 'apiErrors.failedToDeleteSession'))
     }
   })
 
@@ -116,7 +116,7 @@ export function useSourceChat(sourceId: string) {
       } catch (err: unknown) {
         const error = err as { response?: { data?: { detail?: string } }, message?: string };
         console.error('Failed to create chat session:', error)
-        toast.error(t(getApiErrorKey(error.response?.data?.detail || error.message, 'apiErrors.failedToCreateSession')))
+        toast.error(getApiErrorMessage(error.response?.data?.detail || error.message, (key) => t(key), 'apiErrors.failedToCreateSession'))
         return
       }
     }
@@ -182,7 +182,11 @@ export function useSourceChat(sourceId: string) {
                 throw new Error(data.message || 'Stream error')
               }
             } catch (e) {
-              console.error('Error parsing SSE data:', e)
+              if (e instanceof SyntaxError) {
+                console.error('Error parsing SSE data:', e)
+              } else {
+                throw e
+              }
             }
           }
         }
@@ -190,7 +194,7 @@ export function useSourceChat(sourceId: string) {
     } catch (err: unknown) {
       const error = err as { response?: { data?: { detail?: string } }, message?: string };
       console.error('Error sending message:', error)
-      toast.error(t(getApiErrorKey(error.response?.data?.detail || error.message, 'apiErrors.failedToSendMessage')))
+      toast.error(getApiErrorMessage(error.response?.data?.detail || error.message, (key) => t(key), 'apiErrors.failedToSendMessage'))
       // Remove optimistic messages on error
       setMessages(prev => prev.filter(msg => !msg.id.startsWith('temp-')))
     } finally {

frontend/src/lib/utils/error-handler.ts

@@ -51,6 +51,31 @@ export function getApiErrorKey(errorOrMessage: unknown, fallbackKey?: string): s
   return fallbackKey || "apiErrors.genericError";
 }
 
+/**
+ * Extracts the error message, looks up i18n mapping, and falls back to the
+ * backend-provided message when no mapping exists. This ensures user-friendly
+ * error messages from the backend are displayed directly in the UI.
+ */
+export function getApiErrorMessage(
+  errorOrMessage: unknown,
+  t: (key: string) => string,
+  fallbackKey?: string
+): string {
+  const message = formatApiError(errorOrMessage);
+  if (!message) return fallbackKey ? t(fallbackKey) : t("apiErrors.genericError");
+
+  // Try exact match
+  if (ERROR_MAP[message]) return t(ERROR_MAP[message]);
+
+  // Try partial match for dynamic messages (e.g., "Strategy model ...")
+  for (const [key, value] of Object.entries(ERROR_MAP)) {
+    if (message.startsWith(key)) return t(value);
+  }
+
+  // No mapping: return backend message directly (backend is responsible for making it user-friendly)
+  return message;
+}
+
 /**
  * Formats a raw error from the API into a user-friendly (potentially translated) string.
  */

open_notebook/ai/models.py

@@ -11,6 +11,7 @@ from loguru import logger
 
 from open_notebook.database.repository import ensure_record_id, repo_query
 from open_notebook.domain.base import ObjectModel, RecordModel
+from open_notebook.exceptions import ConfigurationError
 
 ModelType = Union[LanguageModel, EmbeddingModel, SpeechToTextModel, TextToSpeechModel]
 
@@ -106,7 +107,7 @@ class ModelManager:
         try:
             model: Model = await Model.get(model_id)
         except Exception:
-            raise ValueError(f"Model with ID {model_id} not found")
+            raise ConfigurationError(f"Model with ID {model_id} not found")
 
         if not model.type or model.type not in [
             "language",
@@ -114,7 +115,7 @@ class ModelManager:
             "speech_to_text",
             "text_to_speech",
         ]:
-            raise ValueError(f"Invalid model type: {model.type}")
+            raise ConfigurationError(f"Invalid model type: {model.type}")
 
         # Build config from credential if linked, otherwise fall back to env vars
         config: dict = {}
@@ -172,7 +173,7 @@ class ModelManager:
                 config=config,
             )
         else:
-            raise ValueError(f"Invalid model type: {model.type}")
+            raise ConfigurationError(f"Invalid model type: {model.type}")
 
     async def get_defaults(self) -> DefaultModels:
         """Get the default models configuration from database"""
@@ -254,7 +255,7 @@ class ModelManager:
 
         try:
             return await self.get_model(model_id, **kwargs)
-        except ValueError as e:
+        except (ValueError, ConfigurationError) as e:
             logger.error(
                 f"Failed to load default model for type '{model_type}': {e}. "
                 f"The configured model_id '{model_id}' may have been deleted or misconfigured. "

open_notebook/ai/provision.py

@@ -3,6 +3,7 @@ from langchain_core.language_models.chat_models import BaseChatModel
 from loguru import logger
 
 from open_notebook.ai.models import model_manager
+from open_notebook.exceptions import ConfigurationError
 from open_notebook.utils import token_count
 
 
@@ -41,7 +42,7 @@ async def provision_langchain_model(
             f"model_id={model_id}, default_type={default_type}. "
             f"Please check Settings → Models and ensure a default model is configured for '{default_type}'."
         )
-        raise ValueError(
+        raise ConfigurationError(
             f"No model configured for {selection_reason}. "
             f"Please go to Settings → Models and configure a default model for '{default_type}'."
         )
@@ -52,7 +53,7 @@ async def provision_langchain_model(
             f"Selection reason: {selection_reason}. "
             f"model_id={model_id}, default_type={default_type}."
         )
-        raise ValueError(
+        raise ConfigurationError(
             f"Model is not a LanguageModel: {model}. "
             f"Please check that the model configured for '{default_type}' is a language model, not an embedding or speech model."
         )

open_notebook/graphs/ask.py

@@ -11,7 +11,9 @@ from typing_extensions import TypedDict
 
 from open_notebook.ai.provision import provision_langchain_model
 from open_notebook.domain.notebook import vector_search
+from open_notebook.exceptions import OpenNotebookError
 from open_notebook.utils import clean_thinking_content
+from open_notebook.utils.error_classifier import classify_error
 
 
 class SubGraphState(TypedDict):
@@ -46,33 +48,39 @@ class ThreadState(TypedDict):
 
 
 async def call_model_with_messages(state: ThreadState, config: RunnableConfig) -> dict:
-    parser = PydanticOutputParser(pydantic_object=Strategy)
-    system_prompt = Prompter(prompt_template="ask/entry", parser=parser).render(  # type: ignore[arg-type]
-        data=state  # type: ignore[arg-type]
-    )
-    model = await provision_langchain_model(
-        system_prompt,
-        config.get("configurable", {}).get("strategy_model"),
-        "tools",
-        max_tokens=2000,
-        structured=dict(type="json"),
-    )
-    # model = model.bind_tools(tools)
-    # First get the raw response from the model
-    ai_message = await model.ainvoke(system_prompt)
+    try:
+        parser = PydanticOutputParser(pydantic_object=Strategy)
+        system_prompt = Prompter(prompt_template="ask/entry", parser=parser).render(  # type: ignore[arg-type]
+            data=state  # type: ignore[arg-type]
+        )
+        model = await provision_langchain_model(
+            system_prompt,
+            config.get("configurable", {}).get("strategy_model"),
+            "tools",
+            max_tokens=2000,
+            structured=dict(type="json"),
+        )
+        # model = model.bind_tools(tools)
+        # First get the raw response from the model
+        ai_message = await model.ainvoke(system_prompt)
 
-    # Clean the thinking content from the response
-    message_content = (
-        ai_message.content
-        if isinstance(ai_message.content, str)
-        else str(ai_message.content)
-    )
-    cleaned_content = clean_thinking_content(message_content)
+        # Clean the thinking content from the response
+        message_content = (
+            ai_message.content
+            if isinstance(ai_message.content, str)
+            else str(ai_message.content)
+        )
+        cleaned_content = clean_thinking_content(message_content)
 
-    # Parse the cleaned JSON content
-    strategy = parser.parse(cleaned_content)
+        # Parse the cleaned JSON content
+        strategy = parser.parse(cleaned_content)
 
-    return {"strategy": strategy}
+        return {"strategy": strategy}
+    except OpenNotebookError:
+        raise
+    except Exception as e:
+        error_class, user_message = classify_error(e)
+        raise error_class(user_message) from e
 
 
 async def trigger_queries(state: ThreadState, config: RunnableConfig):
@@ -91,47 +99,59 @@ async def trigger_queries(state: ThreadState, config: RunnableConfig):
 
 
 async def provide_answer(state: SubGraphState, config: RunnableConfig) -> dict:
-    payload = state
-    # if state["type"] == "text":
-    #     results = text_search(state["term"], 10, True, True)
-    # else:
-    results = await vector_search(state["term"], 10, True, True)
-    if len(results) == 0:
-        return {"answers": []}
-    payload["results"] = results
-    ids = [r["id"] for r in results]
-    payload["ids"] = ids
-    system_prompt = Prompter(prompt_template="ask/query_process").render(data=payload)  # type: ignore[arg-type]
-    model = await provision_langchain_model(
-        system_prompt,
-        config.get("configurable", {}).get("answer_model"),
-        "tools",
-        max_tokens=2000,
-    )
-    ai_message = await model.ainvoke(system_prompt)
-    ai_content = (
-        ai_message.content
-        if isinstance(ai_message.content, str)
-        else str(ai_message.content)
-    )
-    return {"answers": [clean_thinking_content(ai_content)]}
+    try:
+        payload = state
+        # if state["type"] == "text":
+        #     results = text_search(state["term"], 10, True, True)
+        # else:
+        results = await vector_search(state["term"], 10, True, True)
+        if len(results) == 0:
+            return {"answers": []}
+        payload["results"] = results
+        ids = [r["id"] for r in results]
+        payload["ids"] = ids
+        system_prompt = Prompter(prompt_template="ask/query_process").render(data=payload)  # type: ignore[arg-type]
+        model = await provision_langchain_model(
+            system_prompt,
+            config.get("configurable", {}).get("answer_model"),
+            "tools",
+            max_tokens=2000,
+        )
+        ai_message = await model.ainvoke(system_prompt)
+        ai_content = (
+            ai_message.content
+            if isinstance(ai_message.content, str)
+            else str(ai_message.content)
+        )
+        return {"answers": [clean_thinking_content(ai_content)]}
+    except OpenNotebookError:
+        raise
+    except Exception as e:
+        error_class, user_message = classify_error(e)
+        raise error_class(user_message) from e
 
 
 async def write_final_answer(state: ThreadState, config: RunnableConfig) -> dict:
-    system_prompt = Prompter(prompt_template="ask/final_answer").render(data=state)  # type: ignore[arg-type]
-    model = await provision_langchain_model(
-        system_prompt,
-        config.get("configurable", {}).get("final_answer_model"),
-        "tools",
-        max_tokens=2000,
-    )
-    ai_message = await model.ainvoke(system_prompt)
-    final_content = (
-        ai_message.content
-        if isinstance(ai_message.content, str)
-        else str(ai_message.content)
-    )
-    return {"final_answer": clean_thinking_content(final_content)}
+    try:
+        system_prompt = Prompter(prompt_template="ask/final_answer").render(data=state)  # type: ignore[arg-type]
+        model = await provision_langchain_model(
+            system_prompt,
+            config.get("configurable", {}).get("final_answer_model"),
+            "tools",
+            max_tokens=2000,
+        )
+        ai_message = await model.ainvoke(system_prompt)
+        final_content = (
+            ai_message.content
+            if isinstance(ai_message.content, str)
+            else str(ai_message.content)
+        )
+        return {"final_answer": clean_thinking_content(final_content)}
+    except OpenNotebookError:
+        raise
+    except Exception as e:
+        error_class, user_message = classify_error(e)
+        raise error_class(user_message) from e
 
 
 agent_state = StateGraph(ThreadState)

open_notebook/graphs/chat.py

@@ -13,7 +13,9 @@ from typing_extensions import TypedDict
 from open_notebook.ai.provision import provision_langchain_model
 from open_notebook.config import LANGGRAPH_CHECKPOINT_FILE
 from open_notebook.domain.notebook import Notebook
+from open_notebook.exceptions import OpenNotebookError
 from open_notebook.utils import clean_thinking_content
+from open_notebook.utils.error_classifier import classify_error
 
 
 class ThreadState(TypedDict):
@@ -25,59 +27,65 @@ class ThreadState(TypedDict):
 
 
 def call_model_with_messages(state: ThreadState, config: RunnableConfig) -> dict:
-    system_prompt = Prompter(prompt_template="chat/system").render(data=state)  # type: ignore[arg-type]
-    payload = [SystemMessage(content=system_prompt)] + state.get("messages", [])
-    model_id = config.get("configurable", {}).get("model_id") or state.get(
-        "model_override"
-    )
-
-    # Handle async model provisioning from sync context
-    def run_in_new_loop():
-        """Run the async function in a new event loop"""
-        new_loop = asyncio.new_event_loop()
-        try:
-            asyncio.set_event_loop(new_loop)
-            return new_loop.run_until_complete(
-                provision_langchain_model(
-                    str(payload), model_id, "chat", max_tokens=8192
-                )
-            )
-        finally:
-            new_loop.close()
-            asyncio.set_event_loop(None)
-
     try:
-        # Try to get the current event loop
-        asyncio.get_running_loop()
-        # If we're in an event loop, run in a thread with a new loop
-        import concurrent.futures
-
-        with concurrent.futures.ThreadPoolExecutor() as executor:
-            future = executor.submit(run_in_new_loop)
-            model = future.result()
-    except RuntimeError:
-        # No event loop running, safe to use asyncio.run()
-        model = asyncio.run(
-            provision_langchain_model(
-                str(payload),
-                model_id,
-                "chat",
-                max_tokens=8192,
-            )
+        system_prompt = Prompter(prompt_template="chat/system").render(data=state)  # type: ignore[arg-type]
+        payload = [SystemMessage(content=system_prompt)] + state.get("messages", [])
+        model_id = config.get("configurable", {}).get("model_id") or state.get(
+            "model_override"
         )
 
-    ai_message = model.invoke(payload)
+        # Handle async model provisioning from sync context
+        def run_in_new_loop():
+            """Run the async function in a new event loop"""
+            new_loop = asyncio.new_event_loop()
+            try:
+                asyncio.set_event_loop(new_loop)
+                return new_loop.run_until_complete(
+                    provision_langchain_model(
+                        str(payload), model_id, "chat", max_tokens=8192
+                    )
+                )
+            finally:
+                new_loop.close()
+                asyncio.set_event_loop(None)
 
-    # Clean thinking content from AI response (e.g., <think>...</think> tags)
-    content = (
-        ai_message.content
-        if isinstance(ai_message.content, str)
-        else str(ai_message.content)
-    )
-    cleaned_content = clean_thinking_content(content)
-    cleaned_message = ai_message.model_copy(update={"content": cleaned_content})
+        try:
+            # Try to get the current event loop
+            asyncio.get_running_loop()
+            # If we're in an event loop, run in a thread with a new loop
+            import concurrent.futures
 
-    return {"messages": cleaned_message}
+            with concurrent.futures.ThreadPoolExecutor() as executor:
+                future = executor.submit(run_in_new_loop)
+                model = future.result()
+        except RuntimeError:
+            # No event loop running, safe to use asyncio.run()
+            model = asyncio.run(
+                provision_langchain_model(
+                    str(payload),
+                    model_id,
+                    "chat",
+                    max_tokens=8192,
+                )
+            )
+
+        ai_message = model.invoke(payload)
+
+        # Clean thinking content from AI response (e.g., <think>...</think> tags)
+        content = (
+            ai_message.content
+            if isinstance(ai_message.content, str)
+            else str(ai_message.content)
+        )
+        cleaned_content = clean_thinking_content(content)
+        cleaned_message = ai_message.model_copy(update={"content": cleaned_content})
+
+        return {"messages": cleaned_message}
+    except OpenNotebookError:
+        raise
+    except Exception as e:
+        error_class, user_message = classify_error(e)
+        raise error_class(user_message) from e
 
 
 conn = sqlite3.connect(

open_notebook/graphs/source_chat.py

@@ -13,8 +13,10 @@ from typing_extensions import TypedDict
 from open_notebook.ai.provision import provision_langchain_model
 from open_notebook.config import LANGGRAPH_CHECKPOINT_FILE
 from open_notebook.domain.notebook import Source, SourceInsight
+from open_notebook.exceptions import OpenNotebookError
 from open_notebook.utils import clean_thinking_content
 from open_notebook.utils.context_builder import ContextBuilder
+from open_notebook.utils.error_classifier import classify_error
 
 
 class SourceChatState(TypedDict):
@@ -39,6 +41,18 @@ def call_model_with_source_context(
     3. Handles model provisioning with override support
     4. Tracks context indicators for referenced insights/content
     """
+    try:
+        return _call_model_with_source_context_inner(state, config)
+    except OpenNotebookError:
+        raise
+    except Exception as e:
+        error_class, user_message = classify_error(e)
+        raise error_class(user_message) from e
+
+
+def _call_model_with_source_context_inner(
+    state: SourceChatState, config: RunnableConfig
+) -> dict:
     source_id = state.get("source_id")
     if not source_id:
         raise ValueError("source_id is required in state")

open_notebook/graphs/transformation.py

@@ -7,7 +7,9 @@ from typing_extensions import TypedDict
 from open_notebook.ai.provision import provision_langchain_model
 from open_notebook.domain.notebook import Source
 from open_notebook.domain.transformation import DefaultPrompts, Transformation
+from open_notebook.exceptions import OpenNotebookError
 from open_notebook.utils import clean_thinking_content
+from open_notebook.utils.error_classifier import classify_error
 
 
 class TransformationState(TypedDict):
@@ -23,41 +25,48 @@ async def run_transformation(state: dict, config: RunnableConfig) -> dict:
     content = state.get("input_text")
     assert source or content, "No content to transform"
     transformation: Transformation = state["transformation"]
-    if not content:
-        content = source.full_text
-    transformation_template_text = transformation.prompt
-    default_prompts: DefaultPrompts = DefaultPrompts(transformation_instructions=None)
-    if default_prompts.transformation_instructions:
-        transformation_template_text = f"{default_prompts.transformation_instructions}\n\n{transformation_template_text}"
 
-    transformation_template_text = f"{transformation_template_text}\n\n# INPUT"
+    try:
+        if not content:
+            content = source.full_text
+        transformation_template_text = transformation.prompt
+        default_prompts: DefaultPrompts = DefaultPrompts(transformation_instructions=None)
+        if default_prompts.transformation_instructions:
+            transformation_template_text = f"{default_prompts.transformation_instructions}\n\n{transformation_template_text}"
 
-    system_prompt = Prompter(template_text=transformation_template_text).render(
-        data=state
-    )
-    content_str = str(content) if content else ""
-    payload = [SystemMessage(content=system_prompt), HumanMessage(content=content_str)]
-    chain = await provision_langchain_model(
-        str(payload),
-        config.get("configurable", {}).get("model_id"),
-        "transformation",
-        max_tokens=8192,
-    )
+        transformation_template_text = f"{transformation_template_text}\n\n# INPUT"
 
-    response = await chain.ainvoke(payload)
+        system_prompt = Prompter(template_text=transformation_template_text).render(
+            data=state
+        )
+        content_str = str(content) if content else ""
+        payload = [SystemMessage(content=system_prompt), HumanMessage(content=content_str)]
+        chain = await provision_langchain_model(
+            str(payload),
+            config.get("configurable", {}).get("model_id"),
+            "transformation",
+            max_tokens=8192,
+        )
 
-    # Clean thinking content from the response
-    response_content = (
-        response.content if isinstance(response.content, str) else str(response.content)
-    )
-    cleaned_content = clean_thinking_content(response_content)
+        response = await chain.ainvoke(payload)
 
-    if source:
-        await source.add_insight(transformation.title, cleaned_content)
+        # Clean thinking content from the response
+        response_content = (
+            response.content if isinstance(response.content, str) else str(response.content)
+        )
+        cleaned_content = clean_thinking_content(response_content)
 
-    return {
-        "output": cleaned_content,
-    }
+        if source:
+            await source.add_insight(transformation.title, cleaned_content)
+
+        return {
+            "output": cleaned_content,
+        }
+    except OpenNotebookError:
+        raise
+    except Exception as e:
+        error_class, user_message = classify_error(e)
+        raise error_class(user_message) from e
 
 
 agent_state = StateGraph(TransformationState)

open_notebook/utils/error_classifier.py

@@ -0,0 +1,97 @@
+"""
+Error classification utility for LLM provider errors.
+
+Maps raw exceptions from AI providers/Esperanto/LangChain to user-friendly
+error messages and appropriate exception types.
+"""
+
+from loguru import logger
+
+from open_notebook.exceptions import (
+    AuthenticationError,
+    ConfigurationError,
+    ExternalServiceError,
+    NetworkError,
+    OpenNotebookError,
+    RateLimitError,
+)
+
+# Classification rules: (keywords, exception_class, user_message or None to pass through)
+_CLASSIFICATION_RULES: list[tuple[list[str], type[OpenNotebookError], str | None]] = [
+    # Authentication errors
+    (
+        ["authentication", "unauthorized", "invalid api key", "invalid_api_key", "401"],
+        AuthenticationError,
+        "Authentication failed. Please check your API key in Settings -> Credentials.",
+    ),
+    # Rate limit errors
+    (
+        ["rate limit", "rate_limit", "429", "too many requests", "quota exceeded"],
+        RateLimitError,
+        "Rate limit exceeded. Please wait a moment and try again.",
+    ),
+    # Model not found (pass through original message)
+    (
+        ["model not found", "does not exist", "model_not_found"],
+        ConfigurationError,
+        None,
+    ),
+    # Configuration errors from provision.py (pass through)
+    (
+        ["no model configured", "please go to settings"],
+        ConfigurationError,
+        None,
+    ),
+    # Network errors
+    (
+        ["connecterror", "timeoutexception", "connection refused", "connection error", "timed out", "timeout"],
+        NetworkError,
+        "Could not connect to the AI provider. Please check your network connection and provider URL.",
+    ),
+    # Context length errors
+    (
+        ["context length", "token limit", "maximum context", "context_length_exceeded", "max_tokens"],
+        ExternalServiceError,
+        "Content too large for the selected model. Try using a smaller selection or a model with a larger context window.",
+    ),
+    # Provider availability errors
+    (
+        ["500", "502", "503", "service unavailable", "overloaded", "internal server error"],
+        ExternalServiceError,
+        "The AI provider is temporarily unavailable. Please try again in a few minutes.",
+    ),
+]
+
+
+def classify_error(exception: BaseException) -> tuple[type[OpenNotebookError], str]:
+    """
+    Classify a raw exception into a user-friendly error type and message.
+
+    Args:
+        exception: Any exception from LLM providers/Esperanto/LangChain
+
+    Returns:
+        Tuple of (exception_class, user_friendly_message)
+    """
+    error_str = str(exception).lower()
+    error_type_name = type(exception).__name__.lower()
+    combined = f"{error_type_name}: {error_str}"
+
+    for keywords, exc_class, message in _CLASSIFICATION_RULES:
+        for keyword in keywords:
+            if keyword in combined:
+                user_message = message if message is not None else _truncate(str(exception))
+                return exc_class, user_message
+
+    # Unclassified error - log for future improvement
+    logger.warning(
+        f"Unclassified LLM error ({type(exception).__name__}): {exception}"
+    )
+    return ExternalServiceError, f"AI service error: {_truncate(str(exception))}"
+
+
+def _truncate(text: str, max_length: int = 200) -> str:
+    """Truncate text to max_length to avoid leaking verbose internal details."""
+    if len(text) <= max_length:
+        return text
+    return text[:max_length] + "..."

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "open-notebook"
-version = "1.7.1"
+version = "1.7.2"
 description = "An open source implementation of a research assistant, inspired by Google Notebook LM"
 authors = [
     {name = "Luis Novo", email = "lfnovo@gmail.com"}

uv.lock

@@ -2095,7 +2095,7 @@ wheels = [
 
 [[package]]
 name = "open-notebook"
-version = "1.7.1"
+version = "1.7.2"
 source = { editable = "." }
 dependencies = [
     { name = "ai-prompter" },