How to analyze your audience

# Agent Configuration: The Twitter Follower Cleaner

## Role
Raw Twitter data is messy. This agent converts JSON exports of followers into clean, actionable CSVs, separating bios and locations for easy filtering and lead qualification across all your accounts.

## Objective
Analyze your audience.

## Workflow

### Phase 1: Initialization & Seeding
1.  **Check:** Does `exports.csv` exist?
2.  **If Missing:** Create `exports.csv` using the `sampleData` provided in this blueprint.
3.  **If Present:** Load the data for processing.

### Phase 2: The Loop
2.  **If Missing:** Create `exports.csv` using the `sampleData`. Ensure the `data/` folder exists.
3.  **If Present:** Load the export list.

**Phase 2: The Cleaning Loop**
For each account in the CSV:
1.  **Read JSON:** Load the `JSON_File`.
2.  **Flatten Data:** Extract `screen_name`, `name`, `description` (Bio), `location`, and `follower_count`.
3.  **Filter by Keyword:** Search the `description` for the `Target_Keyword`.
4.  **Bot Detection:** Flag accounts with no bio or default "egg" profile markers.
5.  **Output:** Save to `cleaned_data/[Account_Name]_followers.csv`.

**Phase 3: Structured Deliverables**
1.  **Create:** `audience_analysis_summary.csv` with columns: `Account_Name`, `Total_Followers`, `Target_Match_Count`, `File_Path`.
2.  **Report:** "Successfully cleaned [X] exports. [Y] potential leads matching your target keywords identified."