Marketplace : Added categories to chatflows

2026-06-28 23:01:09 +03:00 · 2024-02-05 15:09:44 -05:00
parent cc6a1f04c6
commit 8990b78e10
46 changed files with 67 additions and 38 deletions
@@ -1,5 +1,6 @@
 {
    "description": "Use OpenAI Function Agent and Chain to automatically decide which API to call, generating url and body request from conversation",
    "categories": "Buffer Memory,ChainTool,API Chain,ChatOpenAI,OpenAI Function Agent,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,5 +1,6 @@
 {
    "description": "Given API docs, agent automatically decide which API to call, generating url and body request from conversation",
    "categories": "Buffer Memory,ChainTool,API Chain,ChatOpenAI,Conversational Agent,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,5 +1,6 @@
 {
    "description": "Output antonym of given user input using few-shot prompt template built with examples",
    "categories": "Few Shot Prompt,ChatOpenAI,LLM Chain,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,5 +1,6 @@
 {
    "description": "Use AutoGPT - Autonomous agent with chain of thoughts for self-guided task completion",
    "categories": "AutoGPT,SERP Tool,File Read/Write,ChatOpenAI,Pinecone,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,5 +1,6 @@
 {
    "description": "Use BabyAGI to create tasks and reprioritize for a given objective",
    "categories": "BabyAGI,ChatOpenAI,Pinecone,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,5 +1,6 @@
 {
    "description": "Analyse and summarize CSV data",
    "categories": "CSV Agent,ChatOpenAI,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,5 +1,6 @@
 {
    "description": "Engage with data sources such as YouTube Transcripts, Google, and more through intelligent Q&A interactions",
    "categories": "Memory Vector Store,SearchAPI,ChatOpenAI,Conversational Retrieval QA Chain,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,5 +1,6 @@
 {
    "description": "Use ChatGPT Plugins within LangChain abstractions with GET and POST Tools",
    "categories": "ChatGPT Plugin,HTTP GET/POST,ChatOpenAI,MRKL Agent,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,5 +1,6 @@
 {
    "description": "Use Anthropic Claude with 200k context window to ingest whole document for QnA",
    "categories": "Buffer Memory,Prompt Template,Conversation Chain,ChatAnthropic,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,5 +1,6 @@
 {
    "description": "Answer question based on retrieved documents (context) with built-in memory to remember conversation using LlamaIndex",
    "categories": "Text File,Prompt Template,ChatOpenAI,Conversation Chain,Pinecone,LlamaIndex,Redis",
    "badge": "NEW",
    "nodes": [
        {
@@ -1,5 +1,6 @@
 {
    "description": "A conversational agent for a chat model which utilize chat specific prompts",
    "categories": "Calculator Tool,Buffer Memory,SerpAPI,ChatOpenAI,Conversational Agent,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,5 +1,6 @@
 {
    "description": "Agent optimized for vector retrieval during conversation and answering questions based on previous dialogue.",
    "categories": "Retriever Tool,Buffer Memory,ChatOpenAI,Conversational Retrieval Agent, Pinecone,Langchain",
    "badge": "POPULAR",
    "nodes": [
        {
@@ -1,5 +1,6 @@
 {
    "description": "Text file QnA using conversational retrieval QA chain",
    "categories": "TextFile,ChatOpenAI,Conversational Retrieval QA Chain,Pinecone,Langchain",
    "badge": "POPULAR",
    "nodes": [
        {
@@ -1,5 +1,6 @@
 {
    "description": "Flowise Docs Github QnA using conversational retrieval QA chain",
    "categories": "Memory Vector Store,Github Loader,ChatOpenAI,Conversational Retrieval QA Chain,Langchain",
    "badge": "POPULAR",
    "nodes": [
        {
@@ -1,5 +1,6 @@
 {
    "description": "Simple LLM Chain using HuggingFace Inference API on falcon-7b-instruct model",
    "categories": "HuggingFace,LLM Chain,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,5 +1,6 @@
 {
    "description": "Split flows based on if else condition",
    "categories": "IfElse Function,ChatOpenAI,OpenAI,LLM Chain,Langchain",
    "badge": "new",
    "nodes": [
        {
@@ -1,6 +1,7 @@
 {
    "description": "Generate image using Replicate Stability text-to-image generative AI model",
    "badge": "NEW",
    "categories": "Replicate,ChatOpenAI,LLM Chain,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,6 +1,7 @@
 {
    "description": "Detect text that could generate harmful output and prevent it from being sent to the language model",
    "badge": "NEW",
    "categories": "Moderation,ChatOpenAI,LLM Chain,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,6 +1,7 @@
 {
    "description": "Return response as a list (array) instead of a string/text",
    "badge": "NEW",
    "categories": "CSV Output Parser,ChatOpenAI,LLM Chain,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,6 +1,7 @@
 {
    "description": "QnA chain using Ollama local LLM, LocalAI embedding model, and Faiss local vector store",
    "badge": "POPULAR",
    "categories": "Text File,ChatOllama,Conversational Retrieval QA Chain,Faiss,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,5 +1,6 @@
 {
    "description": "Use long term memory like Zep to differentiate conversations between users with sessionId",
    "categories": "ChatOpenAI,Conversational Retrieval QA Chain,Zep Memory,Qdrant,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,5 +1,6 @@
 {
    "description": "Upsert multiple files with metadata and filter by it using conversational retrieval QA chain",
    "categories": "Text File,PDF File,ChatOpenAI,Conversational Retrieval QA Chain,Pinecone,Langchain",
    "badge": "POPULAR",
    "nodes": [
        {
@@ -1,5 +1,6 @@
 {
    "description": "A chain that automatically picks an appropriate prompt from multiple prompts",
    "categories": "ChatOpenAI,Multi Prompt Chain,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,5 +1,6 @@
 {
    "description": "A chain that automatically picks an appropriate retriever from multiple different vector databases",
    "categories": "ChatOpenAI,Multi Retrieval QA Chain,Pinecone,Chroma,Supabase,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,5 +1,6 @@
 {
    "description": "Use the agent to choose between multiple different vector databases, with the ability to use other tools",
    "categories": "Buffer Memory,ChatOpenAI,Chain Tool,Retrieval QA Chain,Redis,Faiss,Conversational Agent,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,5 +1,6 @@
 {
    "description": "An agent that uses OpenAI's Function Calling functionality to pick the tool and args to call",
    "categories": "Buffer Memory,Custom Tool, SerpAPI,OpenAI Function,Calculator Tool,ChatOpenAI,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,5 +1,6 @@
 {
    "description": "OpenAI Assistant that has instructions and can leverage models, tools, and knowledge to respond to user queries",
    "categories": "Custom Tool, SerpAPI,OpenAI Assistant,Calculator Tool,Langchain",
    "badge": "NEW",
    "nodes": [
        {
@@ -1,5 +1,6 @@
 {
    "description": "Use chat history to rephrase user question, and answer the rephrased question using retrieved docs from vector store",
    "categories": "ChatOpenAI,LLM Chain,SingleStore,Langchain",
    "badge": "POPULAR",
    "nodes": [
        {
@@ -1,5 +1,6 @@
 {
    "description": "Use output from a chain as prompt for another chain",
    "categories": "Custom Tool,OpenAI,LLM Chain,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,5 +1,6 @@
 {
    "description": "Stateless query engine designed to answer question over your data using LlamaIndex",
    "categories": "ChatAnthropic,Compact and Refine,Pinecone,LlamaIndex",
    "badge": "NEW",
    "nodes": [
        {
@@ -1,5 +1,6 @@
 {
    "description": "An agent that uses ReAct logic to decide what action to take",
    "categories": "Calculator Tool,SerpAPI,ChatOpenAI,MRKL Agent,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,5 +1,6 @@
 {
    "description": "Use Replicate API that runs Llama 13b v2 model with LLMChain",
    "categories": "Replicate,LLM Chain,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,5 +1,6 @@
 {
    "description": "Answer questions over a SQL database",
    "categories": "ChatOpenAI,Sql Database Chain,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,5 +1,6 @@
 {
    "description": "Manually construct prompts to query a SQL database",
    "categories": "IfElse Function,Variable Set/Get,Custom JS Function,ChatOpenAI,LLM Chain,Langchain",
    "badge": "new",
    "nodes": [
        {
@@ -1,5 +1,6 @@
 {
    "description": "Simple chat engine to handle back and forth conversations using LlamaIndex",
    "categories": "BufferMemory,AzureChatOpenAI,LlamaIndex",
    "badge": "NEW",
    "nodes": [
        {
@@ -1,5 +1,6 @@
 {
    "description": "Basic example of Conversation Chain with built-in memory - works exactly like ChatGPT",
    "categories": "Buffer Memory,ChatOpenAI,Conversation Chain,Langchain",
    "badge": "POPULAR",
    "nodes": [
        {
@@ -1,5 +1,6 @@
 {
    "description": "Basic example of stateless (no memory) LLM Chain with a Prompt Template and LLM Model",
    "categories": "OpenAI,LLM Chain,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,5 +1,6 @@
 {
    "description": "Return response as a specified JSON structure instead of a string/text",
    "categories": "Structured Output Parser,ChatOpenAI,LLM Chain,Langchain",
    "badge": "NEW",
    "nodes": [
        {
@@ -1,5 +1,6 @@
 {
    "description": "Breaks down query into sub questions for each relevant data source, then combine into final response",
    "categories": "Sub Question Query Engine,Sticky Note,QueryEngine Tool,Compact and Refine,ChatOpenAI,Pinecone,LlamaIndex",
    "badge": "NEW",
    "nodes": [
        {
@@ -1,5 +1,6 @@
 {
    "description": "Language translation using LLM Chain with a Chat Prompt Template and Chat Model",
    "categories": "Chat Prompt Template,ChatOpenAI,LLM Chain,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,4 +1,6 @@
 {
    "description": "QA chain for Vectara",
    "categories": "Vectara QA Chain,Vectara,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,5 +1,6 @@
 {
    "description": "Conversational Agent with ability to visit a website and extract information",
    "categories": "Buffer Memory,Web Browser,ChatOpenAI,Conversational Agent,Langchain",
    "nodes": [
        {
            "width": 300,
@@ -1,5 +1,6 @@
 {
    "description": "Scrape web pages for QnA with long term memory Motorhead and return source documents",
    "categories": "HtmlToMarkdown,Cheerio Web Scraper,ChatOpenAI,Redis,Pinecone,Langchain",
    "badge": "POPULAR",
    "nodes": [
        {
@@ -1237,6 +1237,7 @@ export class App {
                    templateName: file.split('.json')[0],
                    flowData: fileData.toString(),
                    badge: fileDataObj?.badge,
                    categories: fileDataObj?.categories,
                    type: 'Chatflow',
                    description: fileDataObj?.description || ''
                }
@@ -1253,6 +1254,7 @@ export class App {
                    ...fileDataObj,
                    id: index,
                    type: 'Tool',
                    categories: '',
                    templateName: file.split('.json')[0]
                }
                templates.push(template)
@@ -103,45 +103,27 @@ export const MarketplaceTable = ({ data, images, filterFunction, filterByBadge,
                                        </Typography>
                                    </TableCell>
                                    <TableCell key='2'>
-                                        {row.type === 'Chatflow' && images[row.id] && (
+                                        <div
-                                            <div
+                                            style={{
-                                                style={{
+                                                display: 'flex',
-                                                    display: 'flex',
+                                                flexDirection: 'row',
-                                                    flexDirection: 'row',
+                                                flexWrap: 'wrap',
-                                                    flexWrap: 'wrap',
+                                                marginTop: 5
-                                                    marginTop: 5
+                                            }}
-                                                }}
+                                        >
-                                            >
+                                            {row.categories &&
-                                                {images[row.id]
+                                                row.categories
-                                                    .slice(0, images[row.id].length > 5 ? 5 : images[row.id].length)
+                                                    .split(',')
-                                                    .map((img) => (
+                                                    .map((tag, index) => (
-                                                        <div
+                                                        <Chip
-                                                            key={img}
+                                                            variant='outlined'
-                                                            style={{
+                                                            key={index}
-                                                                width: 35,
+                                                            size='small'
-                                                                height: 35,
+                                                            label={tag.toUpperCase()}
-                                                                marginRight: 5,
+                                                            style={{ marginRight: 3, marginBottom: 3 }}
-                                                                borderRadius: '50%',
+                                                        />
                                                                backgroundColor: 'white',
                                                                marginTop: 5
                                                            }}
                                                        >
                                                            <img
                                                                style={{ width: '100%', height: '100%', padding: 5, objectFit: 'contain' }}
                                                                alt=''
                                                                src={img}
                                                            />
                                                        </div>
                                                    ))}
-                                                {images[row.id].length > 5 && (
+                                        </div>
                                                    <Typography
                                                        sx={{ alignItems: 'center', display: 'flex', fontSize: '.8rem', fontWeight: 200 }}
                                                    >
                                                        + {images[row.id].length - 5} More
                                                    </Typography>
                                                )}
                                            </div>
                                        )}
                                    </TableCell>
                                    <TableCell key='3'>
                                        <Typography>
@@ -129,6 +129,7 @@ const Marketplace = () => {
    function filterFlows(data) {
        return (
            data.categories?.toLowerCase().indexOf(search.toLowerCase()) > -1 ||
            data.templateName.toLowerCase().indexOf(search.toLowerCase()) > -1 ||
            (data.description && data.description.toLowerCase().indexOf(search.toLowerCase()) > -1)
        )