LibreChat/packages/data-provider/src/schemas.ts
Danny Avila a0291ed155
🚧 chore: merge latest dev build to main repo (#3844)
* agents - phase 1 (#30)

* chore: copy assistant files

* feat: frontend and data-provider

* feat: backend get endpoint test

* fix(MessageEndpointIcon): switched to AgentName and AgentAvatar

* fix: small fixes

* fix: agent endpoint config

* fix: show Agent Builder

* chore: install agentus

* chore: initial scaffolding for agents

* fix: updated Assistant logic to Agent Logic for some Agent components

* WIP first pass, demo of agent package

* WIP: initial backend infra for agents

* fix: agent list error

* wip: agents routing

* chore: Refactor useSSE hook to handle different data events

* wip: correctly emit events

* chore: Update @librechat/agentus npm dependency to version 1.0.9

* remove comment

* first pass: streaming agent text

* chore: Remove @librechat/agentus root-level workspace npm dependency

* feat: Agent Schema and Model

* fix: content handling fixes

* fix: content message save

* WIP: new content data

* fix: run step issue with tool calls

* chore: Update @librechat/agentus npm dependency to version 1.1.5

* feat: update controller and agent routes

* wip: initial backend tool and tool error handling support

* wip: tool chunks

* chore: Update @librechat/agentus npm dependency to version 1.1.7

* chore: update tool_call typing, add test conditions and logs

* fix: create agent

* fix: create agent

* first pass: render completed content parts

* fix: remove logging, fix step handler typing

* chore: Update @librechat/agentus npm dependency to version 1.1.9

* refactor: cleanup maps on unmount

* chore: Update BaseClient.js to safely count tokens for string, number, and boolean values

* fix: support subsequent messages with tool_calls

* chore: export order

* fix: select agent

* fix: tool call types and handling

* chore: switch to anthropic for testing

* fix: AgentSelect

* refactor: experimental: OpenAIClient to use array for intermediateReply

* fix(useSSE): revert old condition for streaming legacy client tokens

* fix: lint

* revert `agent_id` to `id`

* chore: update localization keys for agent-related components

* feat: zod schema handling for actions

* refactor(actions): if no params, no zodSchema

* chore: Update @librechat/agentus npm dependency to version 1.2.1

* feat: first pass, actions

* refactor: empty schema for actions without params

* feat: Update createRun function to accept additional options

* fix: message payload formatting; feat: add more client options

* fix: ToolCall component rendering when action has no args but has output

* refactor(ToolCall): allow non-stringy args

* WIP: first pass, correctly formatted tool_calls between providers

* refactor: Remove duplicate import of 'roles' module

* refactor: Exclude 'vite.config.ts' from TypeScript compilation

* refactor: fix agent related types
> - no need to use endpoint/model fields for identifying agent metadata
> - add `provider` distinction for agent-configured 'endpoint'
- no need for agent-endpoint map
- reduce complexity of tools as functions into tools as string[]
- fix types related to above changes
- reduce unnecessary variables for queries/mutations and corresponding react-query keys

* refactor: Add tools and tool_kwargs fields to agent schema

* refactor: Remove unused code and update dependencies

* refactor: Update updateAgentHandler to use req.body directly

* refactor: Update AgentSelect component to use localized hooks

* refactor: Update agent schema to include tools and provider fields

* refactor(AgentPanel): add scrollbar gutter, add provider field to form, fix agent schema required values

* refactor: Update AgentSwitcher component to use selectedAgentId instead of selectedAgent

* refactor: Update AgentPanel component to include alternateName import and defaultAgentFormValues

* refactor(SelectDropDown): allow setting value as option while still supporting legacy usage (string values only)

* refactor: SelectDropdown changes - Only necessary when the available values are objects with label/value fields and the selected value is expected to be a string.

* refactor: TypeError issues and handle provider as option

* feat: Add placeholder for provider selection in AgentPanel component

* refactor: Update agent schema to include author and provider fields

* fix: show expected 'create agent' placeholder when creating agent

* chore: fix localization strings, hide capabilities form for now

* chore: typing

* refactor: import order and use compact agents schema for now

* chore: typing

* refactor: Update AgentForm type to use AgentCapabilities

* fix agent form agent selection issues

* feat: responsive agent selection

* fix: Handle cancelled fetch in useSelectAgent hook

* fix: reset agent form on accordion close/open

* feat: Add agent_id to default conversation for agents endpoint

* feat: agents endpoint request handling

* refactor: reset conversation model on agent select

* refactor: add `additional_instructions` to conversation schema, organize other fields

* chore: casing

* chore: types

* refactor(loadAgentTools): explicitly pass agent_id, do not pass `model` to loadAgentTools for now, load action sets by agent_id

* WIP: initial draft of real agent client initialization

* WIP: first pass, anthropic agent requests

* feat: remember last selected agent

* feat: openai and azure connected

* fix: prioritize agent model for runs unless an explicit override model is passed from client

* feat: Agent Actions

* fix: save agent id to convo

* feat: model panel (#29)

* feat: model panel

* bring back comments

* fix: method still null

* fix: AgentPanel FormContext

* feat: add more parameters

* fix: style issues; refactor: Agent Controller

* fix: cherry-pick

* fix: Update AgentAvatar component to use AssistantIcon instead of BrainCircuit

* feat: OGDialog for delete agent; feat(assistant): update Agent types, introduced `model_parameters`

* feat: icon and general `model_parameters` update

* feat: use react-hook-form better

* fix: agent builder form reset issue when switching panels

* refactor: modularize agent builder form

---------

Co-authored-by: Danny Avila <danny@librechat.ai>

* fix: AgentPanel and ModelPanel type issues and use `useFormContext` and `watch` instead of `methods` directly and `useWatch`.

* fix: tool call issues due to invalid input (anthropic) of empty string

* fix: handle empty text in Part component

---------

Co-authored-by: Marco Beretta <81851188+berry-13@users.noreply.github.com>

* refactor: remove form ModelPanel and fixed nested ternary expressions in AgentConfig

* fix: Model Parameters not saved correctly

* refactor: remove console log

* feat: avatar upload and get for Agents (#36)

Co-authored-by: Marco Beretta <81851188+berry-13@users.noreply.github.com>

* chore: update to public package

* fix: typing, optional chaining

* fix: cursor not showing for content parts

* chore: conditionally enable agents

* ci: fix azure test

* ci: fix frontend tests, fix eslint api

* refactor: Remove unused errorContentPart variable

* continue of the agent message PR (#40)

* last fixes

* fix: agentMap

* pr merge test  (#41)

* fix: model icon not fetching correctly

* remove console logs

* feat: agent name

* refactor: pass documentsMap as a prop to allow re-render of assistant form

* refactor: pass documentsMap as a prop to allow re-render of assistant form

* chore: Bump version to 0.7.419

* fix: TypeError: Cannot read properties of undefined (reading 'id')

* refactor: update AgentSwitcher component to use ControlCombobox instead of Combobox

---------

Co-authored-by: Marco Beretta <81851188+berry-13@users.noreply.github.com>
2024-08-31 16:33:51 -04:00

1172 lines
30 KiB
TypeScript

import { z } from 'zod';
import { Tools } from './types/assistants';
import type { TMessageContentParts, FunctionTool, FunctionToolCall } from './types/assistants';
import type { TFile } from './types/files';
export const isUUID = z.string().uuid();
export enum AuthType {
OVERRIDE_AUTH = 'override_auth',
USER_PROVIDED = 'user_provided',
SYSTEM_DEFINED = 'SYSTEM_DEFINED',
}
export const authTypeSchema = z.nativeEnum(AuthType);
export enum EModelEndpoint {
azureOpenAI = 'azureOpenAI',
openAI = 'openAI',
bingAI = 'bingAI',
chatGPTBrowser = 'chatGPTBrowser',
google = 'google',
gptPlugins = 'gptPlugins',
anthropic = 'anthropic',
assistants = 'assistants',
azureAssistants = 'azureAssistants',
agents = 'agents',
custom = 'custom',
}
export type AssistantsEndpoint = EModelEndpoint.assistants | EModelEndpoint.azureAssistants;
export const isAssistantsEndpoint = (endpoint?: AssistantsEndpoint | null | string): boolean => {
if (!endpoint) {
return false;
}
return endpoint.toLowerCase().endsWith(EModelEndpoint.assistants);
};
export type AgentProvider = Exclude<keyof typeof EModelEndpoint, EModelEndpoint.agents> | string;
export const isAgentsEndpoint = (endpoint?: EModelEndpoint.agents | null | string): boolean => {
if (!endpoint) {
return false;
}
return endpoint === EModelEndpoint.agents;
};
export enum ImageDetail {
low = 'low',
auto = 'auto',
high = 'high',
}
export const imageDetailNumeric = {
[ImageDetail.low]: 0,
[ImageDetail.auto]: 1,
[ImageDetail.high]: 2,
};
export const imageDetailValue = {
0: ImageDetail.low,
1: ImageDetail.auto,
2: ImageDetail.high,
};
export const eImageDetailSchema = z.nativeEnum(ImageDetail);
export const defaultAssistantFormValues = {
assistant: '',
id: '',
name: '',
description: '',
instructions: '',
conversation_starters: [],
model: '',
functions: [],
code_interpreter: false,
image_vision: false,
retrieval: false,
};
export const defaultAgentFormValues = {
agent: {},
id: '',
name: '',
description: '',
instructions: '',
model: '',
model_parameters: {},
tools: [],
provider: {},
code_interpreter: false,
image_vision: false,
retrieval: false,
};
export const ImageVisionTool: FunctionTool = {
type: Tools.function,
[Tools.function]: {
name: 'image_vision',
description: 'Get detailed text descriptions for all current image attachments.',
parameters: {
type: 'object',
properties: {},
required: [],
},
},
};
export const isImageVisionTool = (tool: FunctionTool | FunctionToolCall) =>
tool.type === 'function' && tool.function?.name === ImageVisionTool.function?.name;
export const openAISettings = {
model: {
default: 'gpt-4o',
},
temperature: {
min: 0,
max: 1,
step: 0.01,
default: 1,
},
top_p: {
min: 0,
max: 1,
step: 0.01,
default: 1,
},
presence_penalty: {
min: 0,
max: 2,
step: 0.01,
default: 0,
},
frequency_penalty: {
min: 0,
max: 2,
step: 0.01,
default: 0,
},
resendFiles: {
default: true,
},
maxContextTokens: {
default: undefined,
},
max_tokens: {
default: undefined,
},
imageDetail: {
default: ImageDetail.auto,
min: 0,
max: 2,
step: 1,
},
};
export const googleSettings = {
model: {
default: 'gemini-1.5-flash-latest',
},
maxOutputTokens: {
min: 1,
max: 8192,
step: 1,
default: 8192,
},
temperature: {
min: 0,
max: 2,
step: 0.01,
default: 1,
},
topP: {
min: 0,
max: 1,
step: 0.01,
default: 0.95,
},
topK: {
min: 1,
max: 40,
step: 0.01,
default: 40,
},
};
const ANTHROPIC_MAX_OUTPUT = 8192;
const LEGACY_ANTHROPIC_MAX_OUTPUT = 4096;
export const anthropicSettings = {
model: {
default: 'claude-3-5-sonnet-20240620',
},
temperature: {
min: 0,
max: 1,
step: 0.01,
default: 1,
},
promptCache: {
default: true,
},
maxOutputTokens: {
min: 1,
max: ANTHROPIC_MAX_OUTPUT,
step: 1,
default: ANTHROPIC_MAX_OUTPUT,
reset: (modelName: string) => {
if (modelName.includes('claude-3-5-sonnet')) {
return ANTHROPIC_MAX_OUTPUT;
}
return 4096;
},
set: (value: number, modelName: string) => {
if (!modelName.includes('claude-3-5-sonnet') && value > LEGACY_ANTHROPIC_MAX_OUTPUT) {
return LEGACY_ANTHROPIC_MAX_OUTPUT;
}
return value;
},
},
topP: {
min: 0,
max: 1,
step: 0.01,
default: 0.7,
},
topK: {
min: 1,
max: 40,
step: 1,
default: 5,
},
resendFiles: {
default: true,
},
maxContextTokens: {
default: undefined,
},
legacy: {
maxOutputTokens: {
min: 1,
max: LEGACY_ANTHROPIC_MAX_OUTPUT,
step: 1,
default: LEGACY_ANTHROPIC_MAX_OUTPUT,
},
},
};
export const agentsSettings = {
model: {
default: 'gpt-3.5-turbo-test',
},
temperature: {
min: 0,
max: 1,
step: 0.01,
default: 1,
},
top_p: {
min: 0,
max: 1,
step: 0.01,
default: 1,
},
presence_penalty: {
min: 0,
max: 2,
step: 0.01,
default: 0,
},
frequency_penalty: {
min: 0,
max: 2,
step: 0.01,
default: 0,
},
resendFiles: {
default: true,
},
maxContextTokens: {
default: undefined,
},
max_tokens: {
default: undefined,
},
imageDetail: {
default: ImageDetail.auto,
},
};
export const endpointSettings = {
[EModelEndpoint.openAI]: openAISettings,
[EModelEndpoint.google]: googleSettings,
[EModelEndpoint.anthropic]: anthropicSettings,
[EModelEndpoint.agents]: agentsSettings,
};
const google = endpointSettings[EModelEndpoint.google];
export const eModelEndpointSchema = z.nativeEnum(EModelEndpoint);
export const extendedModelEndpointSchema = z.union([eModelEndpointSchema, z.string()]);
export const tPluginAuthConfigSchema = z.object({
authField: z.string(),
label: z.string(),
description: z.string(),
});
export type TPluginAuthConfig = z.infer<typeof tPluginAuthConfigSchema>;
export const tPluginSchema = z.object({
name: z.string(),
pluginKey: z.string(),
description: z.string(),
icon: z.string(),
authConfig: z.array(tPluginAuthConfigSchema),
authenticated: z.boolean().optional(),
isButton: z.boolean().optional(),
});
export type TPlugin = z.infer<typeof tPluginSchema>;
export type TInput = {
inputStr: string;
};
export type TResPlugin = {
plugin: string;
input: string;
thought: string;
loading?: boolean;
outputs?: string;
latest?: string;
inputs?: TInput[];
};
export const tExampleSchema = z.object({
input: z.object({
content: z.string(),
}),
output: z.object({
content: z.string(),
}),
});
export type TExample = z.infer<typeof tExampleSchema>;
export enum EAgent {
functions = 'functions',
classic = 'classic',
}
export const agentOptionSettings = {
model: {
default: 'gpt-4o-mini',
},
temperature: {
min: 0,
max: 1,
step: 0.01,
default: 0,
},
agent: {
default: EAgent.functions,
options: [EAgent.functions, EAgent.classic],
},
skipCompletion: {
default: true,
},
};
export const eAgentOptionsSchema = z.nativeEnum(EAgent);
export const tAgentOptionsSchema = z.object({
agent: z.string().default(EAgent.functions),
skipCompletion: z.boolean().default(agentOptionSettings.skipCompletion.default),
model: z.string(),
temperature: z.number().default(agentOptionSettings.temperature.default),
});
export const tMessageSchema = z.object({
messageId: z.string(),
endpoint: z.string().optional(),
clientId: z.string().nullable().optional(),
conversationId: z.string().nullable(),
parentMessageId: z.string().nullable(),
responseMessageId: z.string().nullable().optional(),
overrideParentMessageId: z.string().nullable().optional(),
bg: z.string().nullable().optional(),
model: z.string().nullable().optional(),
title: z.string().nullable().or(z.literal('New Chat')).default('New Chat'),
sender: z.string(),
text: z.string(),
generation: z.string().nullable().optional(),
isEdited: z.boolean().optional(),
isCreatedByUser: z.boolean(),
error: z.boolean(),
createdAt: z
.string()
.optional()
.default(() => new Date().toISOString()),
updatedAt: z
.string()
.optional()
.default(() => new Date().toISOString()),
current: z.boolean().optional(),
unfinished: z.boolean().optional(),
searchResult: z.boolean().optional(),
finish_reason: z.string().optional(),
/* assistant */
thread_id: z.string().optional(),
/* frontend components */
iconURL: z.string().optional(),
});
export type TMessage = z.input<typeof tMessageSchema> & {
children?: TMessage[];
plugin?: TResPlugin | null;
plugins?: TResPlugin[];
content?: TMessageContentParts[];
files?: Partial<TFile>[];
depth?: number;
siblingIndex?: number;
};
export const coerceNumber = z.union([z.number(), z.string()]).transform((val) => {
if (typeof val === 'string') {
return val.trim() === '' ? undefined : parseFloat(val);
}
return val;
});
export const tConversationSchema = z.object({
conversationId: z.string().nullable(),
endpoint: eModelEndpointSchema.nullable(),
endpointType: eModelEndpointSchema.optional(),
title: z.string().nullable().or(z.literal('New Chat')).default('New Chat'),
user: z.string().optional(),
messages: z.array(z.string()).optional(),
tools: z.union([z.array(tPluginSchema), z.array(z.string())]).optional(),
modelLabel: z.string().nullable().optional(),
userLabel: z.string().optional(),
model: z.string().nullable().optional(),
promptPrefix: z.string().nullable().optional(),
temperature: z.number().optional(),
topP: z.number().optional(),
topK: z.number().optional(),
top_p: z.number().optional(),
frequency_penalty: z.number().optional(),
presence_penalty: z.number().optional(),
parentMessageId: z.string().optional(),
maxOutputTokens: z.number().optional(),
maxContextTokens: coerceNumber.optional(),
max_tokens: coerceNumber.optional(),
/* Anthropic */
promptCache: z.boolean().optional(),
/* artifacts */
artifacts: z.string().optional(),
/* google */
context: z.string().nullable().optional(),
examples: z.array(tExampleSchema).optional(),
/* DB */
tags: z.array(z.string()).optional(),
createdAt: z.string(),
updatedAt: z.string(),
/* Files */
file_ids: z.array(z.string()).optional(),
/* vision */
resendFiles: z.boolean().optional(),
imageDetail: eImageDetailSchema.optional(),
/* assistant */
assistant_id: z.string().optional(),
/* agents */
agent_id: z.string().optional(),
/* assistant + agents */
instructions: z.string().optional(),
additional_instructions: z.string().optional(),
/** Used to overwrite active conversation settings when saving a Preset */
presetOverride: z.record(z.unknown()).optional(),
stop: z.array(z.string()).optional(),
/* frontend components */
iconURL: z.string().optional(),
greeting: z.string().optional(),
spec: z.string().optional(),
/*
Deprecated fields
*/
/** @deprecated */
suggestions: z.array(z.string()).optional(),
/** @deprecated */
systemMessage: z.string().nullable().optional(),
/** @deprecated */
jailbreak: z.boolean().optional(),
/** @deprecated */
jailbreakConversationId: z.string().nullable().optional(),
/** @deprecated */
conversationSignature: z.string().nullable().optional(),
/** @deprecated */
clientId: z.string().nullable().optional(),
/** @deprecated */
invocationId: z.number().nullable().optional(),
/** @deprecated */
toneStyle: z.string().nullable().optional(),
/** @deprecated */
resendImages: z.boolean().optional(),
/** @deprecated */
agentOptions: tAgentOptionsSchema.nullable().optional(),
/** @deprecated Prefer `modelLabel` over `chatGptLabel` */
chatGptLabel: z.string().nullable().optional(),
});
export const tPresetSchema = tConversationSchema
.omit({
conversationId: true,
createdAt: true,
updatedAt: true,
title: true,
})
.merge(
z.object({
conversationId: z.string().nullable().optional(),
presetId: z.string().nullable().optional(),
title: z.string().nullable().optional(),
defaultPreset: z.boolean().optional(),
order: z.number().optional(),
endpoint: extendedModelEndpointSchema.nullable(),
}),
);
export const tConvoUpdateSchema = tConversationSchema.merge(
z.object({
endpoint: extendedModelEndpointSchema.nullable(),
}),
);
export const tPresetUpdateSchema = tConversationSchema.merge(
z.object({
endpoint: extendedModelEndpointSchema.nullable(),
}),
);
export type TPreset = z.infer<typeof tPresetSchema>;
export type TSetOption = (
param: number | string,
) => (newValue: number | string | boolean | string[] | Partial<TPreset>) => void;
export type TConversation = z.infer<typeof tConversationSchema> & {
presetOverride?: Partial<TPreset>;
};
export const tSharedLinkSchema = z.object({
conversationId: z.string(),
shareId: z.string(),
messages: z.array(z.string()),
isAnonymous: z.boolean(),
isPublic: z.boolean(),
isVisible: z.boolean(),
title: z.string(),
createdAt: z.string(),
updatedAt: z.string(),
});
export type TSharedLink = z.infer<typeof tSharedLinkSchema>;
export const tConversationTagSchema = z.object({
_id: z.string(),
user: z.string(),
tag: z.string(),
description: z.string().optional(),
createdAt: z.string(),
updatedAt: z.string(),
count: z.number(),
position: z.number(),
});
export type TConversationTag = z.infer<typeof tConversationTagSchema>;
export const openAISchema = tConversationSchema
.pick({
model: true,
modelLabel: true,
chatGptLabel: true,
promptPrefix: true,
temperature: true,
top_p: true,
presence_penalty: true,
frequency_penalty: true,
resendFiles: true,
artifacts: true,
imageDetail: true,
stop: true,
iconURL: true,
greeting: true,
spec: true,
maxContextTokens: true,
max_tokens: true,
})
.transform((obj) => {
const result = {
...obj,
model: obj.model ?? openAISettings.model.default,
chatGptLabel: obj.chatGptLabel ?? obj.modelLabel ?? null,
promptPrefix: obj.promptPrefix ?? null,
temperature: obj.temperature ?? openAISettings.temperature.default,
top_p: obj.top_p ?? openAISettings.top_p.default,
presence_penalty: obj.presence_penalty ?? openAISettings.presence_penalty.default,
frequency_penalty: obj.frequency_penalty ?? openAISettings.frequency_penalty.default,
resendFiles:
typeof obj.resendFiles === 'boolean' ? obj.resendFiles : openAISettings.resendFiles.default,
imageDetail: obj.imageDetail ?? openAISettings.imageDetail.default,
stop: obj.stop ?? undefined,
iconURL: obj.iconURL ?? undefined,
greeting: obj.greeting ?? undefined,
spec: obj.spec ?? undefined,
maxContextTokens: obj.maxContextTokens ?? undefined,
max_tokens: obj.max_tokens ?? undefined,
};
if (obj.modelLabel) {
result.modelLabel = null;
}
return result;
})
.catch(() => ({
model: openAISettings.model.default,
chatGptLabel: null,
promptPrefix: null,
temperature: openAISettings.temperature.default,
top_p: openAISettings.top_p.default,
presence_penalty: openAISettings.presence_penalty.default,
frequency_penalty: openAISettings.frequency_penalty.default,
resendFiles: openAISettings.resendFiles.default,
imageDetail: openAISettings.imageDetail.default,
stop: undefined,
iconURL: undefined,
greeting: undefined,
spec: undefined,
maxContextTokens: undefined,
max_tokens: undefined,
}));
export const googleSchema = tConversationSchema
.pick({
model: true,
modelLabel: true,
promptPrefix: true,
examples: true,
temperature: true,
maxOutputTokens: true,
artifacts: true,
topP: true,
topK: true,
iconURL: true,
greeting: true,
spec: true,
maxContextTokens: true,
})
.transform((obj) => {
return {
...obj,
model: obj.model ?? google.model.default,
modelLabel: obj.modelLabel ?? null,
promptPrefix: obj.promptPrefix ?? null,
examples: obj.examples ?? [{ input: { content: '' }, output: { content: '' } }],
temperature: obj.temperature ?? google.temperature.default,
maxOutputTokens: obj.maxOutputTokens ?? google.maxOutputTokens.default,
topP: obj.topP ?? google.topP.default,
topK: obj.topK ?? google.topK.default,
iconURL: obj.iconURL ?? undefined,
greeting: obj.greeting ?? undefined,
spec: obj.spec ?? undefined,
maxContextTokens: obj.maxContextTokens ?? undefined,
};
})
.catch(() => ({
model: google.model.default,
modelLabel: null,
promptPrefix: null,
examples: [{ input: { content: '' }, output: { content: '' } }],
temperature: google.temperature.default,
maxOutputTokens: google.maxOutputTokens.default,
topP: google.topP.default,
topK: google.topK.default,
iconURL: undefined,
greeting: undefined,
spec: undefined,
maxContextTokens: undefined,
}));
export const bingAISchema = tConversationSchema
.pick({
jailbreak: true,
systemMessage: true,
context: true,
toneStyle: true,
jailbreakConversationId: true,
conversationSignature: true,
clientId: true,
invocationId: true,
})
.transform((obj) => ({
...obj,
model: '',
jailbreak: obj.jailbreak ?? false,
systemMessage: obj.systemMessage ?? null,
context: obj.context ?? null,
toneStyle: obj.toneStyle ?? 'creative',
jailbreakConversationId: obj.jailbreakConversationId ?? null,
conversationSignature: obj.conversationSignature ?? null,
clientId: obj.clientId ?? null,
invocationId: obj.invocationId ?? 1,
}))
.catch(() => ({
model: '',
jailbreak: false,
systemMessage: null,
context: null,
toneStyle: 'creative',
jailbreakConversationId: null,
conversationSignature: null,
clientId: null,
invocationId: 1,
}));
export const anthropicSchema = tConversationSchema
.pick({
model: true,
modelLabel: true,
promptPrefix: true,
temperature: true,
maxOutputTokens: true,
topP: true,
topK: true,
resendFiles: true,
promptCache: true,
artifacts: true,
iconURL: true,
greeting: true,
spec: true,
maxContextTokens: true,
})
.transform((obj) => {
const model = obj.model ?? anthropicSettings.model.default;
return {
...obj,
model,
modelLabel: obj.modelLabel ?? null,
promptPrefix: obj.promptPrefix ?? null,
temperature: obj.temperature ?? anthropicSettings.temperature.default,
maxOutputTokens: obj.maxOutputTokens ?? anthropicSettings.maxOutputTokens.reset(model),
topP: obj.topP ?? anthropicSettings.topP.default,
topK: obj.topK ?? anthropicSettings.topK.default,
promptCache:
typeof obj.promptCache === 'boolean'
? obj.promptCache
: anthropicSettings.promptCache.default,
resendFiles:
typeof obj.resendFiles === 'boolean'
? obj.resendFiles
: anthropicSettings.resendFiles.default,
iconURL: obj.iconURL ?? undefined,
greeting: obj.greeting ?? undefined,
spec: obj.spec ?? undefined,
maxContextTokens: obj.maxContextTokens ?? anthropicSettings.maxContextTokens.default,
};
})
.catch(() => ({
model: anthropicSettings.model.default,
modelLabel: null,
promptPrefix: null,
temperature: anthropicSettings.temperature.default,
maxOutputTokens: anthropicSettings.maxOutputTokens.default,
topP: anthropicSettings.topP.default,
topK: anthropicSettings.topK.default,
resendFiles: anthropicSettings.resendFiles.default,
promptCache: anthropicSettings.promptCache.default,
iconURL: undefined,
greeting: undefined,
spec: undefined,
maxContextTokens: anthropicSettings.maxContextTokens.default,
}));
export const chatGPTBrowserSchema = tConversationSchema
.pick({
model: true,
})
.transform((obj) => ({
...obj,
model: obj.model ?? 'text-davinci-002-render-sha',
}))
.catch(() => ({
model: 'text-davinci-002-render-sha',
}));
export const gptPluginsSchema = tConversationSchema
.pick({
model: true,
modelLabel: true,
chatGptLabel: true,
promptPrefix: true,
temperature: true,
artifacts: true,
top_p: true,
presence_penalty: true,
frequency_penalty: true,
tools: true,
agentOptions: true,
iconURL: true,
greeting: true,
spec: true,
maxContextTokens: true,
})
.transform((obj) => {
const result = {
...obj,
model: obj.model ?? 'gpt-3.5-turbo',
chatGptLabel: obj.chatGptLabel ?? obj.modelLabel ?? null,
promptPrefix: obj.promptPrefix ?? null,
temperature: obj.temperature ?? 0.8,
top_p: obj.top_p ?? 1,
presence_penalty: obj.presence_penalty ?? 0,
frequency_penalty: obj.frequency_penalty ?? 0,
tools: obj.tools ?? [],
agentOptions: obj.agentOptions ?? {
agent: EAgent.functions,
skipCompletion: true,
model: 'gpt-3.5-turbo',
temperature: 0,
},
iconURL: obj.iconURL ?? undefined,
greeting: obj.greeting ?? undefined,
spec: obj.spec ?? undefined,
maxContextTokens: obj.maxContextTokens ?? undefined,
};
if (obj.modelLabel) {
result.modelLabel = null;
}
return result;
})
.catch(() => ({
model: 'gpt-3.5-turbo',
chatGptLabel: null,
promptPrefix: null,
temperature: 0.8,
top_p: 1,
presence_penalty: 0,
frequency_penalty: 0,
tools: [],
agentOptions: {
agent: EAgent.functions,
skipCompletion: true,
model: 'gpt-3.5-turbo',
temperature: 0,
},
iconURL: undefined,
greeting: undefined,
spec: undefined,
maxContextTokens: undefined,
}));
export function removeNullishValues<T extends object>(obj: T): T {
const newObj: Partial<T> = { ...obj };
(Object.keys(newObj) as Array<keyof T>).forEach((key) => {
if (newObj[key] === undefined || newObj[key] === null || newObj[key] === '') {
delete newObj[key];
}
});
return newObj as T;
}
export const assistantSchema = tConversationSchema
.pick({
model: true,
assistant_id: true,
instructions: true,
artifacts: true,
promptPrefix: true,
iconURL: true,
greeting: true,
spec: true,
})
.transform((obj) => ({
...obj,
model: obj.model ?? openAISettings.model.default,
assistant_id: obj.assistant_id ?? undefined,
instructions: obj.instructions ?? undefined,
promptPrefix: obj.promptPrefix ?? null,
iconURL: obj.iconURL ?? undefined,
greeting: obj.greeting ?? undefined,
spec: obj.spec ?? undefined,
}))
.catch(() => ({
model: openAISettings.model.default,
assistant_id: undefined,
instructions: undefined,
promptPrefix: null,
iconURL: undefined,
greeting: undefined,
spec: undefined,
}));
export const compactAssistantSchema = tConversationSchema
.pick({
model: true,
assistant_id: true,
instructions: true,
promptPrefix: true,
artifacts: true,
iconURL: true,
greeting: true,
spec: true,
})
// will change after adding temperature
.transform(removeNullishValues)
.catch(() => ({}));
export const agentsSchema = tConversationSchema
.pick({
model: true,
modelLabel: true,
temperature: true,
top_p: true,
presence_penalty: true,
frequency_penalty: true,
resendFiles: true,
imageDetail: true,
agent_id: true,
instructions: true,
promptPrefix: true,
iconURL: true,
greeting: true,
maxContextTokens: true,
})
.transform((obj) => ({
...obj,
model: obj.model ?? agentsSettings.model.default,
modelLabel: obj.modelLabel ?? null,
temperature: obj.temperature ?? 1,
top_p: obj.top_p ?? 1,
presence_penalty: obj.presence_penalty ?? 0,
frequency_penalty: obj.frequency_penalty ?? 0,
resendFiles:
typeof obj.resendFiles === 'boolean' ? obj.resendFiles : agentsSettings.resendFiles.default,
imageDetail: obj.imageDetail ?? ImageDetail.auto,
agent_id: obj.agent_id ?? undefined,
instructions: obj.instructions ?? undefined,
promptPrefix: obj.promptPrefix ?? null,
iconURL: obj.iconURL ?? undefined,
greeting: obj.greeting ?? undefined,
maxContextTokens: obj.maxContextTokens ?? undefined,
}))
.catch(() => ({
model: agentsSettings.model.default,
modelLabel: null,
temperature: 1,
top_p: 1,
presence_penalty: 0,
frequency_penalty: 0,
resendFiles: agentsSettings.resendFiles.default,
imageDetail: ImageDetail.auto,
agent_id: undefined,
instructions: undefined,
promptPrefix: null,
iconURL: undefined,
greeting: undefined,
maxContextTokens: undefined,
}));
export const compactAgentsSchema = tConversationSchema
.pick({
model: true,
agent_id: true,
instructions: true,
promptPrefix: true,
iconURL: true,
greeting: true,
spec: true,
})
.transform(removeNullishValues)
.catch(() => ({}));
export const compactOpenAISchema = tConversationSchema
.pick({
model: true,
chatGptLabel: true,
promptPrefix: true,
temperature: true,
top_p: true,
presence_penalty: true,
frequency_penalty: true,
resendFiles: true,
artifacts: true,
imageDetail: true,
stop: true,
iconURL: true,
greeting: true,
spec: true,
maxContextTokens: true,
max_tokens: true,
})
.transform((obj: Partial<TConversation>) => {
const newObj: Partial<TConversation> = { ...obj };
if (newObj.temperature === openAISettings.temperature.default) {
delete newObj.temperature;
}
if (newObj.top_p === openAISettings.top_p.default) {
delete newObj.top_p;
}
if (newObj.presence_penalty === openAISettings.presence_penalty.default) {
delete newObj.presence_penalty;
}
if (newObj.frequency_penalty === openAISettings.frequency_penalty.default) {
delete newObj.frequency_penalty;
}
if (newObj.resendFiles === openAISettings.resendFiles.default) {
delete newObj.resendFiles;
}
if (newObj.imageDetail === openAISettings.imageDetail.default) {
delete newObj.imageDetail;
}
return removeNullishValues(newObj);
})
.catch(() => ({}));
export const compactGoogleSchema = tConversationSchema
.pick({
model: true,
modelLabel: true,
promptPrefix: true,
examples: true,
temperature: true,
maxOutputTokens: true,
artifacts: true,
topP: true,
topK: true,
iconURL: true,
greeting: true,
spec: true,
maxContextTokens: true,
})
.transform((obj) => {
const newObj: Partial<TConversation> = { ...obj };
if (newObj.temperature === google.temperature.default) {
delete newObj.temperature;
}
if (newObj.maxOutputTokens === google.maxOutputTokens.default) {
delete newObj.maxOutputTokens;
}
if (newObj.topP === google.topP.default) {
delete newObj.topP;
}
if (newObj.topK === google.topK.default) {
delete newObj.topK;
}
return removeNullishValues(newObj);
})
.catch(() => ({}));
export const compactAnthropicSchema = tConversationSchema
.pick({
model: true,
modelLabel: true,
promptPrefix: true,
temperature: true,
maxOutputTokens: true,
topP: true,
topK: true,
resendFiles: true,
promptCache: true,
artifacts: true,
iconURL: true,
greeting: true,
spec: true,
maxContextTokens: true,
})
.transform((obj) => {
const newObj: Partial<TConversation> = { ...obj };
if (newObj.temperature === anthropicSettings.temperature.default) {
delete newObj.temperature;
}
if (newObj.maxOutputTokens === anthropicSettings.legacy.maxOutputTokens.default) {
delete newObj.maxOutputTokens;
}
if (newObj.topP === anthropicSettings.topP.default) {
delete newObj.topP;
}
if (newObj.topK === anthropicSettings.topK.default) {
delete newObj.topK;
}
if (newObj.resendFiles === anthropicSettings.resendFiles.default) {
delete newObj.resendFiles;
}
if (newObj.promptCache === anthropicSettings.promptCache.default) {
delete newObj.promptCache;
}
return removeNullishValues(newObj);
})
.catch(() => ({}));
export const compactChatGPTSchema = tConversationSchema
.pick({
model: true,
})
.transform((obj) => {
const newObj: Partial<TConversation> = { ...obj };
return removeNullishValues(newObj);
})
.catch(() => ({}));
export const compactPluginsSchema = tConversationSchema
.pick({
model: true,
chatGptLabel: true,
promptPrefix: true,
temperature: true,
top_p: true,
presence_penalty: true,
frequency_penalty: true,
tools: true,
agentOptions: true,
iconURL: true,
greeting: true,
spec: true,
maxContextTokens: true,
})
.transform((obj) => {
const newObj: Partial<TConversation> = { ...obj };
if (newObj.chatGptLabel === null) {
delete newObj.chatGptLabel;
}
if (newObj.promptPrefix === null) {
delete newObj.promptPrefix;
}
if (newObj.temperature === 0.8) {
delete newObj.temperature;
}
if (newObj.top_p === 1) {
delete newObj.top_p;
}
if (newObj.presence_penalty === 0) {
delete newObj.presence_penalty;
}
if (newObj.frequency_penalty === 0) {
delete newObj.frequency_penalty;
}
if (newObj.tools?.length === 0) {
delete newObj.tools;
}
if (
newObj.agentOptions &&
newObj.agentOptions.agent === EAgent.functions &&
newObj.agentOptions.skipCompletion === true &&
newObj.agentOptions.model === 'gpt-3.5-turbo' &&
newObj.agentOptions.temperature === 0
) {
delete newObj.agentOptions;
}
return removeNullishValues(newObj);
})
.catch(() => ({}));