Business List from Location Search

Create me an app that can run from any browser. The app will search the internet using google. Here is an example: Find businesses of any kind using Google followed by location. Example city or zip code which is input by the user. The finished product will create an excel spreatsheet that lists the city, business name, phone number, street address and zip. Also website URL if available. The app would work like this. Create me a spreadsheet for plumbers in Dallas Texas. The App would then create and exportan excell spreadsheet containing what I have just listed. The App would have my company name which is Crescent Technical. Make it easy to use. No graphics needed. Make it simle, just create the excel spreadsheets and fill in data as found. Do not make it comples as you seem to want to do. Just search for a company type using an input of city or zip and fill in the fields as requested, then creat the spreadsheet.

First, pre-entered info...

final-output2

The final output is a single Excel file (.xlsx) listing each business with these columns: • City • Business Name • Phone Number • Street Address • ZIP Code • Website URL (if available) This looks very good

subagents-extracted2

OVERVIEW OF SUBAGENTS: "Google Search Query Agent" • Purpose: Takes the user's inputs (business type + location) and searches the web for relevant business listings. • Skill Used: • #224 (Oracle Ask A Question) – Queries live web sources and returns textual results about businesses. • Input: "Find [BUSINESS_TYPE] in [LOCATION] with name, address, phone, ZIP, website." • Output: Unstructured text detailing businesses found. SUBAGENT FINAL OUTPUT: [raw-business-search-results] SUBAGENT 1: "Google Search Query Agent" • Purpose: Takes the user's inputs (business type + location) and searches the web for relevant business listings. • Skill Used: • #224 (Oracle Ask A Question) – Queries live web sources and returns textual results about businesses. • Input: "Find [BUSINESS_TYPE] in [LOCATION] with name, address, phone, ZIP, website." • Output: Unstructured text detailing businesses found. SUBAGENT FINAL OUTPUT: [raw-business-search-results] SUBAGENT 2: "Business Data Extraction Agent" • Purpose: From the unstructured text, extract each business's details: Business Name, Street Address, City, ZIP, Phone, Website. • Skill Used: • #190 (Write or rewrite text based on instructions) – Parses and rewrites the raw text to isolate the fields. • Input: Text from Subagent 1's output + instructions for how to extract details. • Output: A textual list of businesses, each with {Name, Address, City, ZIP, Phone, Website}. SUBAGENT FINAL OUTPUT: [extracted-business-details] SUBAGENT 3: "Data Structuring Agent" • Purpose: Convert extracted details into a consistent, line-by-line structure (making it ready for CSV). • Skill Used: • #190 (Write or rewrite text based on instructions) – Takes the parsed business data and ensures uniform field order. • Input: Structured text from Subagent 2 + instructions on how to unify fields (comma-separated). • Output: A cleaned, consistently ordered data set, line by line, ready to become CSV. SUBAGENT FINAL OUTPUT: [structured-business-data] SUBAGENT 4: "CSV Generator Agent" • Purpose: Create final CSV text (which can be opened in Excel). • Skill Used: • #190 (Write or rewrite text based on instructions) – Converts the structured text into CSV format with a header row: City, Business Name, Phone Number, Street Address, ZIP Code, Website URL. • Input: Uniform data from Subagent 3 + instructions for CSV formatting. • Output: Plain-text CSV content (e.g., "CrescentTechnical_Business_Search.csv"). SUBAGENT FINAL OUTPUT: [csv-file-data] SUBAGENT 5: "Export Agent" • Purpose: Provide the CSV output to the user, so they can save or open it in Excel. • Implementation: Since no dedicated download skill exists, this final step simply presents the CSV text or link. SUBAGENT FINAL OUTPUT: [final-csv-export]

subagentnumber

[subagentnumber]

Then, agent response...

subagent1refined-new-json-core

G) TASKFLOW SUMMARY Based on the subagent description, here are the core skills/tasks identified: SKILL 1: - ID: #190 - Title task: Query String Construction - Description: Formats the user's business type and location inputs into a standardized search query string for the Oracle skill to process - Input(s) involved: [businessType], [location] - Output token name: [formatted-query-string] SKILL 2: - ID: #224 - Title task: Oracle Web Search Execution - Description: Performs a live web search using the formatted query to find business listings matching the specified criteria, returning unstructured text results containing business details - Input(s) involved: [formatted-query-string] - Output token name: [raw-business-search-results] Note: While Skill #185 was mentioned as an optional alternative for query construction, the core workflow primarily relies on #190 and #224. The taskflow follows this sequence: 1. User inputs are processed using Skill #190 to create a properly formatted search query 2. The formatted query is passed to Skill #224 to perform the actual web search 3. The raw results are returned as unstructured text The final output token [raw-business-search-results] contains the unstructured text data that will be passed to subsequent subagents for parsing and structuring. This represents the complete taskflow for this initial subagent, focusing on data retrieval rather than processing or formatting, which are handled by later subagents in the chain.

subagent1mermaid

<!DOCTYPE html> <html> <head> <title>Mermaid Diagram Example</title> <script src="https://cdnjs.cloudflare.com/ajax/libs/mermaid/9.4.3/mermaid.min.js"></script> </head> <body> <div class="mermaid"> graph TD A[Start] --> B[Input Fields] B --> C[Business Type] B --> D[Location] C --> E[Query Construction] D --> E E --> |Skill 190| F[Format Query String] F --> G[Web Search] G --> |Skill 224| H[Execute Oracle Search] H --> I[Raw Results] I --> J[Unstructured Text Output] J --> K[End Subagent] style A fill:#f9f,stroke:#333,stroke-width:2px style K fill:#f9f,stroke:#333,stroke-width:2px style E fill:#bbf,stroke:#333,stroke-width:2px style G fill:#bbf,stroke:#333,stroke-width:2px </div> <script> mermaid.initialize({ startOnLoad: true }); </script> </body> </html>

Mermaid Image - https://static.aiz.ac/1736201142-mermaid/mermaid1.png

prompts-required

[prompts-required]

subagent1-json-array

[ { "taskID": 190, "semanticTitleOfTask": "Query String Construction - create a Google search query to find business listings", "taskDescription": "This skill constructs a specialized Google search query using the provided business type and location. The generated query is optimized to find relevant business listings and contact details such as company names, addresses, phone numbers, ZIP codes, and websites.", "inputDescription": "This requires two string inputs: businessType (e.g. 'plumbers') and location (e.g. 'Dallas Texas') that are used to construct a specialized search query text.", "inputRequired": ["businessType", "location"], "outputDescription": "A single string that represents a well-formed query to be passed to the Oracle Web Search Execution skill, optimized to locate business listings from online directories and search engines.", "outputName": "formatted-query-string", "promptInstruction": "Create a Google search query to find business listings. Use the business type `businessType` and location `location` to construct a search string that will return business details including company names, addresses, phone numbers, ZIP codes, and websites. The query should be optimized to find business directory listings and contact information. Format the query to maximize relevant results." }, { "taskID": 224, "semanticTitleOfTask": "Oracle Web Search Execution - gather raw business data from the web", "taskDescription": "This skill performs a live web search using the previously constructed query string to find business listings in the specified location. It returns unstructured text containing relevant data such as addresses, phone numbers, ZIP codes, and websites for each listing.", "inputDescription": "This task requires a single string input, the 'formatted-query-string' produced by the Query String Construction skill, to search web sources and return unstructured business listings.", "inputRequired": ["formatted-query-string"], "outputDescription": "Unstructured text containing raw business listings discovered during the web search, including any available contact information, addresses, ZIP codes, or websites.", "outputName": "raw-business-search-results", "promptInstruction": "Search for businesses matching `formatted-query-string`. For each business found, retrieve the following information if available: - Business name - Street address - City - ZIP code - Phone number - Website URL Return all findings as text, including as many relevant businesses as possible within the specified location. Focus on directory listings, business profile pages, and official business websites." } ]

subagent1-json-array-mermaid

<!DOCTYPE html> <html> <head> <title>Mermaid Diagram Example</title> <script src="https://cdnjs.cloudflare.com/ajax/libs/mermaid/9.4.3/mermaid.min.js"></script> </head> <body> <div class="mermaid"> graph TD A[Start] --> B[User Input] B --> C[Extract Business Type] B --> D[Extract Location] C --> E[Query String Construction - Task 190] D --> E E --> F[formatted-query-string] F --> G[Oracle Web Search - Task 224] G --> H[raw-business-search-results] H --> I[Pass to Next Subagent] I --> J[End] subgraph User Interface A B end subgraph Query Construction C D E F end subgraph Web Search G H end subgraph Output I J end </div> <script> mermaid.initialize({ startOnLoad: true }); </script> </body> </html>