mirror of
https://github.com/danny-avila/LibreChat.git
synced 2025-12-17 17:00:15 +01:00
🎉 feat: Code Interpreter API and Agents Release (#4860)
* feat: Code Interpreter API & File Search Agent Uploads chore: add back code files wip: first pass, abstract key dialog refactor: influence checkbox on key changes refactor: update localization keys for 'execute code' to 'run code' wip: run code button refactor: add throwError parameter to loadAuthValues and getUserPluginAuthValue functions feat: first pass, API tool calling fix: handle missing toolId in callTool function and return 404 for non-existent tools feat: show code outputs fix: improve error handling in callTool function and log errors fix: handle potential null value for filepath in attachment destructuring fix: normalize language before rendering and prevent null return fix: add loading indicator in RunCode component while executing code feat: add support for conditional code execution in Markdown components feat: attachments refactor: remove bash fix: pass abort signal to graph/run refactor: debounce and rate limit tool call refactor: increase debounce delay for execute function feat: set code output attachments feat: image attachments refactor: apply message context refactor: pass `partIndex` feat: toolCall schema/model/methods feat: block indexing feat: get tool calls chore: imports chore: typing chore: condense type imports feat: get tool calls fix: block indexing chore: typing refactor: update tool calls mapping to support multiple results fix: add unique key to nav link for rendering wip: first pass, tool call results refactor: update query cache from successful tool call mutation style: improve result switcher styling chore: note on using \`.toObject()\` feat: add agent_id field to conversation schema chore: typing refactor: rename agentMap to agentsMap for consistency feat: Agent Name as chat input placeholder chore: bump agents 📦 chore: update @langchain dependencies to latest versions to match agents package 📦 chore: update @librechat/agents dependency to version 1.8.0 fix: Aborting agent stream removes sender; fix(bedrock): completion removes preset name label refactor: remove direct file parameter to use req.file, add `processAgentFileUpload` for image uploads feat: upload menu feat: prime message_file resources feat: implement conversation access validation in chat route refactor: remove file parameter from processFileUpload and use req.file instead feat: add savedMessageIds set to track saved message IDs in BaseClient, to prevent unnecessary double-write to db feat: prevent duplicate message saves by checking savedMessageIds in AgentController refactor: skip legacy RAG API handling for agents feat: add files field to convoSchema refactor: update request type annotations from Express.Request to ServerRequest in file processing functions feat: track conversation files fix: resendFiles, addPreviousAttachments handling feat: add ID validation for session_id and file_id in download route feat: entity_id for code file uploads/downloads fix: code file edge cases feat: delete related tool calls feat: add stream rate handling for LLM configuration feat: enhance system content with attached file information fix: improve error logging in resource priming function * WIP: PoC, sequential agents WIP: PoC Sequential Agents, first pass content data + bump agents package fix: package-lock WIP: PoC, o1 support, refactor bufferString feat: convertJsonSchemaToZod fix: form issues and schema defining erroneous model fix: max length issue on agent form instructions, limit conversation messages to sequential agents feat: add abort signal support to createRun function and AgentClient feat: PoC, hide prior sequential agent steps fix: update parameter naming from config to metadata in event handlers for clarity, add model to usage data refactor: use only last contentData, track model for usage data chore: bump agents package fix: content parts issue refactor: filter contentParts to include tool calls and relevant indices feat: show function calls refactor: filter context messages to exclude tool calls when no tools are available to the agent fix: ensure tool call content is not undefined in formatMessages feat: add agent_id field to conversationPreset schema feat: hide sequential agents feat: increase upload toast duration to 10 seconds * refactor: tool context handling & update Code API Key Dialog feat: toolContextMap chore: skipSpecs -> useSpecs ci: fix handleTools tests feat: API Key Dialog * feat: Agent Permissions Admin Controls feat: replace label with button for prompt permission toggle feat: update agent permissions feat: enable experimental agents and streamline capability configuration feat: implement access control for agents and enhance endpoint menu items feat: add welcome message for agent selection in localization feat: add agents permission to access control and update version to 0.7.57 * fix: update types in useAssistantListMap and useMentions hooks for better null handling * feat: mention agents * fix: agent tool resource race conditions when deleting agent tool resource files * feat: add error handling for code execution with user feedback * refactor: rename AdminControls to AdminSettings for clarity * style: add gap to button in AdminSettings for improved layout * refactor: separate agent query hooks and check access to enable fetching * fix: remove unused provider from agent initialization options, creates issue with custom endpoints * refactor: remove redundant/deprecated modelOptions from AgentClient processes * chore: update @librechat/agents to version 1.8.5 in package.json and package-lock.json * fix: minor styling issues + agent panel uniformity * fix: agent edge cases when set endpoint is no longer defined * refactor: remove unused cleanup function call from AppService * fix: update link in ApiKeyDialog to point to pricing page * fix: improve type handling and layout calculations in SidePanel component * fix: add missing localization string for agent selection in SidePanel * chore: form styling and localizations for upload filesearch/code interpreter * fix: model selection placeholder logic in AgentConfig component * style: agent capabilities * fix: add localization for provider selection and improve dropdown styling in ModelPanel * refactor: use gpt-4o-mini > gpt-3.5-turbo * fix: agents configuration for loadDefaultInterface and update related tests * feat: DALLE Agents support
This commit is contained in:
Danny Avila
GitHub
parent
affcebd48c
commit
1a815f5e19
189 changed files with 5056 additions and 1815 deletions
|
|
@ -12,9 +12,11 @@ const {
|
|||
Constants,
|
||||
VisionModes,
|
||||
openAISchema,
|
||||
ContentTypes,
|
||||
EModelEndpoint,
|
||||
KnownEndpoints,
|
||||
anthropicSchema,
|
||||
isAgentsEndpoint,
|
||||
bedrockOutputParser,
|
||||
removeNullishValues,
|
||||
} = require('librechat-data-provider');
|
||||
|
|
@ -30,10 +32,10 @@ const {
|
|||
createContextHandlers,
|
||||
} = require('~/app/clients/prompts');
|
||||
const { encodeAndFormat } = require('~/server/services/Files/images/encode');
|
||||
const { getBufferString, HumanMessage } = require('@langchain/core/messages');
|
||||
const Tokenizer = require('~/server/services/Tokenizer');
|
||||
const { spendTokens } = require('~/models/spendTokens');
|
||||
const BaseClient = require('~/app/clients/BaseClient');
|
||||
// const { sleep } = require('~/server/utils');
|
||||
const { createRun } = require('./run');
|
||||
const { logger } = require('~/config');
|
||||
|
||||
|
|
@ -48,6 +50,12 @@ const providerParsers = {
|
|||
|
||||
const legacyContentEndpoints = new Set([KnownEndpoints.groq, KnownEndpoints.deepseek]);
|
||||
|
||||
const noSystemModelRegex = [/\bo1\b/gi];
|
||||
|
||||
// const { processMemory, memoryInstructions } = require('~/server/services/Endpoints/agents/memory');
|
||||
// const { getFormattedMemories } = require('~/models/Memory');
|
||||
// const { getCurrentDateTime } = require('~/utils');
|
||||
|
||||
class AgentClient extends BaseClient {
|
||||
constructor(options = {}) {
|
||||
super(null, options);
|
||||
|
|
@ -62,15 +70,15 @@ class AgentClient extends BaseClient {
|
|||
this.run;
|
||||
|
||||
const {
|
||||
agentConfigs,
|
||||
contentParts,
|
||||
collectedUsage,
|
||||
artifactPromises,
|
||||
maxContextTokens,
|
||||
modelOptions = {},
|
||||
...clientOptions
|
||||
} = options;
|
||||
|
||||
this.modelOptions = modelOptions;
|
||||
this.agentConfigs = agentConfigs;
|
||||
this.maxContextTokens = maxContextTokens;
|
||||
/** @type {MessageContentComplex[]} */
|
||||
this.contentParts = contentParts;
|
||||
|
|
@ -80,6 +88,8 @@ class AgentClient extends BaseClient {
|
|||
this.artifactPromises = artifactPromises;
|
||||
/** @type {AgentClientOptions} */
|
||||
this.options = Object.assign({ endpoint: options.endpoint }, clientOptions);
|
||||
/** @type {string} */
|
||||
this.model = this.options.agent.model_parameters.model;
|
||||
}
|
||||
|
||||
/**
|
||||
|
|
@ -169,7 +179,7 @@ class AgentClient extends BaseClient {
|
|||
: {};
|
||||
|
||||
if (parseOptions) {
|
||||
runOptions = parseOptions(this.modelOptions);
|
||||
runOptions = parseOptions(this.options.agent.model_parameters);
|
||||
}
|
||||
|
||||
return removeNullishValues(
|
||||
|
|
@ -224,7 +234,28 @@ class AgentClient extends BaseClient {
|
|||
let promptTokens;
|
||||
|
||||
/** @type {string} */
|
||||
let systemContent = `${instructions ?? ''}${additional_instructions ?? ''}`;
|
||||
let systemContent = [instructions ?? '', additional_instructions ?? '']
|
||||
.filter(Boolean)
|
||||
.join('\n')
|
||||
.trim();
|
||||
// this.systemMessage = getCurrentDateTime();
|
||||
// const { withKeys, withoutKeys } = await getFormattedMemories({
|
||||
// userId: this.options.req.user.id,
|
||||
// });
|
||||
// processMemory({
|
||||
// userId: this.options.req.user.id,
|
||||
// message: this.options.req.body.text,
|
||||
// parentMessageId,
|
||||
// memory: withKeys,
|
||||
// thread_id: this.conversationId,
|
||||
// }).catch((error) => {
|
||||
// logger.error('Memory Agent failed to process memory', error);
|
||||
// });
|
||||
|
||||
// this.systemMessage += '\n\n' + memoryInstructions;
|
||||
// if (withoutKeys) {
|
||||
// this.systemMessage += `\n\n# Existing memory about the user:\n${withoutKeys}`;
|
||||
// }
|
||||
|
||||
if (this.options.attachments) {
|
||||
const attachments = await this.options.attachments;
|
||||
|
|
@ -245,7 +276,8 @@ class AgentClient extends BaseClient {
|
|||
this.options.attachments = files;
|
||||
}
|
||||
|
||||
if (this.message_file_map) {
|
||||
/** Note: Bedrock uses legacy RAG API handling */
|
||||
if (this.message_file_map && !isAgentsEndpoint(this.options.endpoint)) {
|
||||
this.contextHandlers = createContextHandlers(
|
||||
this.options.req,
|
||||
orderedMessages[orderedMessages.length - 1].text,
|
||||
|
|
@ -319,7 +351,6 @@ class AgentClient extends BaseClient {
|
|||
|
||||
/** @type {sendCompletion} */
|
||||
async sendCompletion(payload, opts = {}) {
|
||||
this.modelOptions.user = this.user;
|
||||
await this.chatCompletion({
|
||||
payload,
|
||||
onProgress: opts.onProgress,
|
||||
|
|
@ -339,10 +370,10 @@ class AgentClient extends BaseClient {
|
|||
await spendTokens(
|
||||
{
|
||||
context,
|
||||
model: model ?? this.modelOptions.model,
|
||||
conversationId: this.conversationId,
|
||||
user: this.user ?? this.options.req.user?.id,
|
||||
endpointTokenConfig: this.options.endpointTokenConfig,
|
||||
model: usage.model ?? model ?? this.model ?? this.options.agent.model_parameters.model,
|
||||
},
|
||||
{ promptTokens: usage.input_tokens, completionTokens: usage.output_tokens },
|
||||
);
|
||||
|
|
@ -457,43 +488,190 @@ class AgentClient extends BaseClient {
|
|||
// });
|
||||
// }
|
||||
|
||||
const run = await createRun({
|
||||
req: this.options.req,
|
||||
agent: this.options.agent,
|
||||
tools: this.options.tools,
|
||||
runId: this.responseMessageId,
|
||||
modelOptions: this.modelOptions,
|
||||
customHandlers: this.options.eventHandlers,
|
||||
});
|
||||
|
||||
const config = {
|
||||
configurable: {
|
||||
thread_id: this.conversationId,
|
||||
last_agent_index: this.agentConfigs?.size ?? 0,
|
||||
hide_sequential_outputs: this.options.agent.hide_sequential_outputs,
|
||||
},
|
||||
signal: abortController.signal,
|
||||
streamMode: 'values',
|
||||
version: 'v2',
|
||||
};
|
||||
|
||||
if (!run) {
|
||||
throw new Error('Failed to create run');
|
||||
}
|
||||
|
||||
this.run = run;
|
||||
|
||||
const messages = formatAgentMessages(payload);
|
||||
const initialMessages = formatAgentMessages(payload);
|
||||
if (legacyContentEndpoints.has(this.options.agent.endpoint)) {
|
||||
formatContentStrings(messages);
|
||||
formatContentStrings(initialMessages);
|
||||
}
|
||||
await run.processStream({ messages }, config, {
|
||||
[Callback.TOOL_ERROR]: (graph, error, toolId) => {
|
||||
logger.error(
|
||||
'[api/server/controllers/agents/client.js #chatCompletion] Tool Error',
|
||||
error,
|
||||
toolId,
|
||||
);
|
||||
},
|
||||
|
||||
/** @type {ReturnType<createRun>} */
|
||||
let run;
|
||||
|
||||
/**
|
||||
*
|
||||
* @param {Agent} agent
|
||||
* @param {BaseMessage[]} messages
|
||||
* @param {number} [i]
|
||||
* @param {TMessageContentParts[]} [contentData]
|
||||
*/
|
||||
const runAgent = async (agent, messages, i = 0, contentData = []) => {
|
||||
config.configurable.model = agent.model_parameters.model;
|
||||
if (i > 0) {
|
||||
this.model = agent.model_parameters.model;
|
||||
}
|
||||
config.configurable.agent_id = agent.id;
|
||||
config.configurable.name = agent.name;
|
||||
config.configurable.agent_index = i;
|
||||
const noSystemMessages = noSystemModelRegex.some((regex) =>
|
||||
agent.model_parameters.model.match(regex),
|
||||
);
|
||||
|
||||
const systemMessage = Object.values(agent.toolContextMap ?? {})
|
||||
.join('\n')
|
||||
.trim();
|
||||
|
||||
let systemContent = [
|
||||
systemMessage,
|
||||
agent.instructions ?? '',
|
||||
i !== 0 ? agent.additional_instructions ?? '' : '',
|
||||
]
|
||||
.join('\n')
|
||||
.trim();
|
||||
|
||||
if (noSystemMessages === true) {
|
||||
agent.instructions = undefined;
|
||||
agent.additional_instructions = undefined;
|
||||
} else {
|
||||
agent.instructions = systemContent;
|
||||
agent.additional_instructions = undefined;
|
||||
}
|
||||
|
||||
if (noSystemMessages === true && systemContent?.length) {
|
||||
let latestMessage = messages.pop().content;
|
||||
if (typeof latestMessage !== 'string') {
|
||||
latestMessage = latestMessage[0].text;
|
||||
}
|
||||
latestMessage = [systemContent, latestMessage].join('\n');
|
||||
messages.push(new HumanMessage(latestMessage));
|
||||
}
|
||||
|
||||
run = await createRun({
|
||||
agent,
|
||||
req: this.options.req,
|
||||
runId: this.responseMessageId,
|
||||
signal: abortController.signal,
|
||||
customHandlers: this.options.eventHandlers,
|
||||
});
|
||||
|
||||
if (!run) {
|
||||
throw new Error('Failed to create run');
|
||||
}
|
||||
|
||||
if (i === 0) {
|
||||
this.run = run;
|
||||
}
|
||||
|
||||
if (contentData.length) {
|
||||
run.Graph.contentData = contentData;
|
||||
}
|
||||
|
||||
await run.processStream({ messages }, config, {
|
||||
keepContent: i !== 0,
|
||||
callbacks: {
|
||||
[Callback.TOOL_ERROR]: (graph, error, toolId) => {
|
||||
logger.error(
|
||||
'[api/server/controllers/agents/client.js #chatCompletion] Tool Error',
|
||||
error,
|
||||
toolId,
|
||||
);
|
||||
},
|
||||
},
|
||||
});
|
||||
};
|
||||
|
||||
await runAgent(this.options.agent, initialMessages);
|
||||
|
||||
let finalContentStart = 0;
|
||||
if (this.agentConfigs && this.agentConfigs.size > 0) {
|
||||
let latestMessage = initialMessages.pop().content;
|
||||
if (typeof latestMessage !== 'string') {
|
||||
latestMessage = latestMessage[0].text;
|
||||
}
|
||||
let i = 1;
|
||||
let runMessages = [];
|
||||
|
||||
const lastFiveMessages = initialMessages.slice(-5);
|
||||
for (const [agentId, agent] of this.agentConfigs) {
|
||||
if (abortController.signal.aborted === true) {
|
||||
break;
|
||||
}
|
||||
const currentRun = await run;
|
||||
|
||||
if (
|
||||
i === this.agentConfigs.size &&
|
||||
config.configurable.hide_sequential_outputs === true
|
||||
) {
|
||||
const content = this.contentParts.filter(
|
||||
(part) => part.type === ContentTypes.TOOL_CALL,
|
||||
);
|
||||
|
||||
this.options.res.write(
|
||||
`event: message\ndata: ${JSON.stringify({
|
||||
event: 'on_content_update',
|
||||
data: {
|
||||
runId: this.responseMessageId,
|
||||
content,
|
||||
},
|
||||
})}\n\n`,
|
||||
);
|
||||
}
|
||||
const _runMessages = currentRun.Graph.getRunMessages();
|
||||
finalContentStart = this.contentParts.length;
|
||||
runMessages = runMessages.concat(_runMessages);
|
||||
const contentData = currentRun.Graph.contentData.slice();
|
||||
const bufferString = getBufferString([new HumanMessage(latestMessage), ...runMessages]);
|
||||
if (i === this.agentConfigs.size) {
|
||||
logger.debug(`SEQUENTIAL AGENTS: Last buffer string:\n${bufferString}`);
|
||||
}
|
||||
try {
|
||||
const contextMessages = [];
|
||||
for (const message of lastFiveMessages) {
|
||||
const messageType = message._getType();
|
||||
if (
|
||||
(!agent.tools || agent.tools.length === 0) &&
|
||||
(messageType === 'tool' || (message.tool_calls?.length ?? 0) > 0)
|
||||
) {
|
||||
continue;
|
||||
}
|
||||
|
||||
contextMessages.push(message);
|
||||
}
|
||||
const currentMessages = [...contextMessages, new HumanMessage(bufferString)];
|
||||
await runAgent(agent, currentMessages, i, contentData);
|
||||
} catch (err) {
|
||||
logger.error(
|
||||
`[api/server/controllers/agents/client.js #chatCompletion] Error running agent ${agentId} (${i})`,
|
||||
err,
|
||||
);
|
||||
}
|
||||
i++;
|
||||
}
|
||||
}
|
||||
|
||||
if (config.configurable.hide_sequential_outputs !== true) {
|
||||
finalContentStart = 0;
|
||||
}
|
||||
|
||||
this.contentParts = this.contentParts.filter((part, index) => {
|
||||
// Include parts that are either:
|
||||
// 1. At or after the finalContentStart index
|
||||
// 2. Of type tool_call
|
||||
// 3. Have tool_call_ids property
|
||||
return (
|
||||
index >= finalContentStart || part.type === ContentTypes.TOOL_CALL || part.tool_call_ids
|
||||
);
|
||||
});
|
||||
|
||||
this.recordCollectedUsage({ context: 'message' }).catch((err) => {
|
||||
logger.error(
|
||||
'[api/server/controllers/agents/client.js #chatCompletion] Error recording collected usage',
|
||||
|
|
@ -586,7 +764,7 @@ class AgentClient extends BaseClient {
|
|||
}
|
||||
|
||||
getEncoding() {
|
||||
return this.modelOptions.model?.includes('gpt-4o') ? 'o200k_base' : 'cl100k_base';
|
||||
return this.model?.includes('gpt-4o') ? 'o200k_base' : 'cl100k_base';
|
||||
}
|
||||
|
||||
/**
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue