LibreChat/api/app/clients/tools/util/handleTools.js
Danny Avila 317a1bd8da
feat: ConversationSummaryBufferMemory (#973)
* refactor: pass model in message edit payload, use encoder in standalone util function

* feat: add summaryBuffer helper

* refactor(api/messages): use new countTokens helper and add auth middleware at top

* wip: ConversationSummaryBufferMemory

* refactor: move pre-generation helpers to prompts dir

* chore: remove console log

* chore: remove test as payload will no longer carry tokenCount

* chore: update getMessagesWithinTokenLimit JSDoc

* refactor: optimize getMessagesForConversation and also break on summary, feat(ci): getMessagesForConversation tests

* refactor(getMessagesForConvo): count '00000000-0000-0000-0000-000000000000' as root message

* chore: add newer model to token map

* fix: condition was point to prop of array instead of message prop

* refactor(BaseClient): use object for refineMessages param, rename 'summary' to 'summaryMessage', add previous_summary
refactor(getMessagesWithinTokenLimit): replace text and tokenCount if should summarize, summary, and summaryTokenCount are present
fix/refactor(handleContextStrategy): use the right comparison length for context diff, and replace payload first message when a summary is present

* chore: log previous_summary if debugging

* refactor(formatMessage): assume if role is defined that it's a valid value

* refactor(getMessagesWithinTokenLimit): remove summary logic
refactor(handleContextStrategy): add usePrevSummary logic in case only summary was pruned
refactor(loadHistory): initial message query will return all ordered messages but keep track of the latest summary
refactor(getMessagesForConversation): use object for single param, edit jsdoc, edit all files using the method
refactor(ChatGPTClient): order messages before buildPrompt is called, TODO: add convoSumBuffMemory logic

* fix: undefined handling and summarizing only when shouldRefineContext is true

* chore(BaseClient): fix test results omitting system role for summaries and test edge case

* chore: export summaryBuffer from index file

* refactor(OpenAIClient/BaseClient): move refineMessages to subclass, implement LLM initialization for summaryBuffer

* feat: add OPENAI_SUMMARIZE to enable summarizing, refactor: rename client prop 'shouldRefineContext' to 'shouldSummarize', change contextStrategy value to 'summarize' from 'refine'

* refactor: rename refineMessages method to summarizeMessages for clarity

* chore: clarify summary future intent in .env.example

* refactor(initializeLLM): handle case for either 'model' or 'modelName' being passed

* feat(gptPlugins): enable summarization for plugins

* refactor(gptPlugins): utilize new initializeLLM method and formatting methods for messages, use payload array for currentMessages and assign pastMessages sooner

* refactor(agents): use ConversationSummaryBufferMemory for both agent types

* refactor(formatMessage): optimize original method for langchain, add helper function for langchain messages, add JSDocs and tests

* refactor(summaryBuffer): add helper to createSummaryBufferMemory, and use new formatting helpers

* fix: forgot to spread formatMessages also took opportunity to pluralize filename

* refactor: pass memory to tools, namely openapi specs. not used and may never be used by new method but added for testing

* ci(formatMessages): add more exhaustive checks for langchain messages

* feat: add debug env var for OpenAI

* chore: delete unnecessary comments

* chore: add extra note about summary feature

* fix: remove tokenCount from payload instructions

* fix: test fail

* fix: only pass instructions to payload when defined or not empty object

* refactor: fromPromptMessages is deprecated, use renamed method fromMessages

* refactor: use 'includes' instead of 'startsWith' for extended OpenRouter compatibility

* fix(PluginsClient.buildPromptBody): handle undefined message strings

* chore: log langchain titling error

* feat: getModelMaxTokens helper

* feat: tokenSplit helper

* feat: summary prompts updated

* fix: optimize _CUT_OFF_SUMMARIZER prompt

* refactor(summaryBuffer): use custom summary prompt, allow prompt to be passed, pass humanPrefix and aiPrefix to memory, along with any future variables, rename messagesToRefine to context

* fix(summaryBuffer): handle edge case where messagesToRefine exceeds summary context,
refactor(BaseClient): allow custom maxContextTokens to be passed to getMessagesWithinTokenLimit, add defined check before unshifting summaryMessage, update shouldSummarize based on this
refactor(OpenAIClient): use getModelMaxTokens, use cut-off message method for summary if no messages were left after pruning

* fix(handleContextStrategy): handle case where incoming prompt is bigger than model context

* chore: rename refinedContent to splitText

* chore: remove unnecessary debug log
2023-09-26 21:02:28 -04:00

263 lines
6.8 KiB
JavaScript

const { getUserPluginAuthValue } = require('../../../../server/services/PluginService');
const { OpenAIEmbeddings } = require('langchain/embeddings/openai');
const { ZapierToolKit } = require('langchain/agents');
const { SerpAPI, ZapierNLAWrapper } = require('langchain/tools');
const { ChatOpenAI } = require('langchain/chat_models/openai');
const { Calculator } = require('langchain/tools/calculator');
const { WebBrowser } = require('langchain/tools/webbrowser');
const {
availableTools,
CodeInterpreter,
AIPluginTool,
GoogleSearchAPI,
WolframAlphaAPI,
StructuredWolfram,
HttpRequestTool,
OpenAICreateImage,
StableDiffusionAPI,
StructuredSD,
AzureCognitiveSearch,
StructuredACS,
E2BTools,
CodeSherpa,
CodeSherpaTools,
CodeBrew,
} = require('../');
const { loadSpecs } = require('./loadSpecs');
const { loadToolSuite } = require('./loadToolSuite');
const getOpenAIKey = async (options, user) => {
let openAIApiKey = options.openAIApiKey ?? process.env.OPENAI_API_KEY;
openAIApiKey = openAIApiKey === 'user_provided' ? null : openAIApiKey;
return openAIApiKey || (await getUserPluginAuthValue(user, 'OPENAI_API_KEY'));
};
const validateTools = async (user, tools = []) => {
try {
const validToolsSet = new Set(tools);
const availableToolsToValidate = availableTools.filter((tool) =>
validToolsSet.has(tool.pluginKey),
);
const validateCredentials = async (authField, toolName) => {
const adminAuth = process.env[authField];
if (adminAuth && adminAuth.length > 0) {
return;
}
const userAuth = await getUserPluginAuthValue(user, authField);
if (userAuth && userAuth.length > 0) {
return;
}
validToolsSet.delete(toolName);
};
for (const tool of availableToolsToValidate) {
if (!tool.authConfig || tool.authConfig.length === 0) {
continue;
}
for (const auth of tool.authConfig) {
await validateCredentials(auth.authField, tool.pluginKey);
}
}
return Array.from(validToolsSet.values());
} catch (err) {
console.log('There was a problem validating tools', err);
throw new Error(err);
}
};
const loadToolWithAuth = async (user, authFields, ToolConstructor, options = {}) => {
return async function () {
let authValues = {};
for (const authField of authFields) {
let authValue = process.env[authField];
if (!authValue) {
authValue = await getUserPluginAuthValue(user, authField);
}
authValues[authField] = authValue;
}
return new ToolConstructor({ ...options, ...authValues });
};
};
const loadTools = async ({
user,
model,
functions = null,
returnMap = false,
tools = [],
options = {},
}) => {
const toolConstructors = {
calculator: Calculator,
codeinterpreter: CodeInterpreter,
google: GoogleSearchAPI,
wolfram: functions ? StructuredWolfram : WolframAlphaAPI,
'dall-e': OpenAICreateImage,
'stable-diffusion': functions ? StructuredSD : StableDiffusionAPI,
'azure-cognitive-search': functions ? StructuredACS : AzureCognitiveSearch,
CodeBrew: CodeBrew,
};
const openAIApiKey = await getOpenAIKey(options, user);
const customConstructors = {
e2b_code_interpreter: async () => {
if (!functions) {
return null;
}
return await loadToolSuite({
pluginKey: 'e2b_code_interpreter',
tools: E2BTools,
user,
options: {
model,
openAIApiKey,
...options,
},
});
},
codesherpa_tools: async () => {
if (!functions) {
return null;
}
return await loadToolSuite({
pluginKey: 'codesherpa_tools',
tools: CodeSherpaTools,
user,
options,
});
},
'web-browser': async () => {
// let openAIApiKey = options.openAIApiKey ?? process.env.OPENAI_API_KEY;
// openAIApiKey = openAIApiKey === 'user_provided' ? null : openAIApiKey;
// openAIApiKey = openAIApiKey || (await getUserPluginAuthValue(user, 'OPENAI_API_KEY'));
const browser = new WebBrowser({ model, embeddings: new OpenAIEmbeddings({ openAIApiKey }) });
browser.description_for_model = browser.description;
return browser;
},
serpapi: async () => {
let apiKey = process.env.SERPAPI_API_KEY;
if (!apiKey) {
apiKey = await getUserPluginAuthValue(user, 'SERPAPI_API_KEY');
}
return new SerpAPI(apiKey, {
location: 'Austin,Texas,United States',
hl: 'en',
gl: 'us',
});
},
zapier: async () => {
let apiKey = process.env.ZAPIER_NLA_API_KEY;
if (!apiKey) {
apiKey = await getUserPluginAuthValue(user, 'ZAPIER_NLA_API_KEY');
}
const zapier = new ZapierNLAWrapper({ apiKey });
return ZapierToolKit.fromZapierNLAWrapper(zapier);
},
plugins: async () => {
return [
new HttpRequestTool(),
await AIPluginTool.fromPluginUrl(
'https://www.klarna.com/.well-known/ai-plugin.json',
new ChatOpenAI({ openAIApiKey: options.openAIApiKey, temperature: 0 }),
),
];
},
};
const requestedTools = {};
if (functions) {
toolConstructors.codesherpa = CodeSherpa;
}
const toolOptions = {
serpapi: { location: 'Austin,Texas,United States', hl: 'en', gl: 'us' },
};
const toolAuthFields = {};
availableTools.forEach((tool) => {
if (customConstructors[tool.pluginKey]) {
return;
}
toolAuthFields[tool.pluginKey] = tool.authConfig.map((auth) => auth.authField);
});
const remainingTools = [];
for (const tool of tools) {
if (customConstructors[tool]) {
requestedTools[tool] = customConstructors[tool];
continue;
}
if (toolConstructors[tool]) {
const options = toolOptions[tool] || {};
const toolInstance = await loadToolWithAuth(
user,
toolAuthFields[tool],
toolConstructors[tool],
options,
);
requestedTools[tool] = toolInstance;
continue;
}
if (functions) {
remainingTools.push(tool);
}
}
let specs = null;
if (functions && remainingTools.length > 0) {
specs = await loadSpecs({
llm: model,
user,
message: options.message,
memory: options.memory,
tools: remainingTools,
map: true,
verbose: false,
});
}
for (const tool of remainingTools) {
if (specs && specs[tool]) {
requestedTools[tool] = specs[tool];
}
}
if (returnMap) {
return requestedTools;
}
// load tools
let result = [];
for (const tool of tools) {
const validTool = requestedTools[tool];
const plugin = await validTool();
if (Array.isArray(plugin)) {
result = [...result, ...plugin];
} else if (plugin) {
result.push(plugin);
}
}
return result;
};
module.exports = {
validateTools,
loadTools,
};