How are you running AnythingLLM?
AnythingLLM desktop app
What happened?
After setting up a workspace and choosing any LLM I can get started chatting in a new thread. however, anything llm is overwriting my modelfile and preset system prompt (when using Ollama) So I go to workshop settings to add a system prompt, change my llm to the one I want and add openai agent llm.
I cannot make any changes in the Workspace LLM Provider or Workspace Agent LLM Provider without causing this error:
Error:
Invalid prisma.workspaces.update() invocation:
{
where: {
id: 1
},
data: {
chatProvider: "ollama",
chatModel: "demicube_tablereader:latest",
chatMode: "chat",
openAiHistory: 20,
openAiPrompt: "Given the following conversation, relevant context, and a follow up question, reply with an answer to the current question the user is asking. Return only your response to the question given the above information following the users instructions as needed.",
queryRefusalResponse: "There is no relevant information in this workspace to answer your query.",
openAiTemp: 0.7,
? name?: String | StringFieldUpdateOperationsInput,
? slug?: String | StringFieldUpdateOperationsInput,
? vectorTag?: String | NullableStringFieldUpdateOperationsInput | Null,
? createdAt?: DateTime | DateTimeFieldUpdateOperationsInput,
? lastUpdatedAt?: DateTime | DateTimeFieldUpdateOperationsInput,
? similarityThreshold?: Float | NullableFloatFieldUpdateOperationsInput | Null,
? topN?: Int | NullableIntFieldUpdateOperationsInput | Null,
? pfpFilename?: String | NullableStringFieldUpdateOperationsInput | Null,
? workspace_users?: workspace_usersUpdateManyWithoutWorkspacesNestedInput,
? documents?: workspace_documentsUpdateManyWithoutWorkspaceNestedInput,
? workspace_suggested_messages?: workspace_suggested_messagesUpdateManyWithoutWorkspaceNestedInput,
? embed_configs?: embed_configsUpdateManyWithoutWorkspaceNestedInput,
? threads?: workspace_threadsUpdateManyWithoutWorkspaceNestedInput
}
}
Unknown argument `chatProvider`. Available options are listed in green.

### Are there known steps to reproduce?
1. set up new workspace
2. try to change any settings in Workspace LLM Provider or Workspace Agent LLM Provider
How are you running AnythingLLM?
AnythingLLM desktop app
What happened?
After setting up a workspace and choosing any LLM I can get started chatting in a new thread. however, anything llm is overwriting my modelfile and preset system prompt (when using Ollama) So I go to workshop settings to add a system prompt, change my llm to the one I want and add openai agent llm.
I cannot make any changes in the Workspace LLM Provider or Workspace Agent LLM Provider without causing this error:
Error:
Invalid
prisma.workspaces.update()invocation:{
where: {
id: 1
},
data: {
chatProvider: "ollama",