automation-script-builder

name	automation-script-builder
description	Build automation scripts for repetitive tasks like file processing, data transformation, scheduling, API calls, and workflow automation. Use when the user needs a script to automate something, process files in batch, schedule tasks, or connect tools. Triggers on "automate this", "write a script for", "batch process", "schedule this", "connect these tools", or any automation request.

HONEST CLASSIFICATION: This is a prompt template, not an executable plugin. OpenClaw injects this as context to guide LLM behavior. It does NOT enforce rules at runtime. Real enforcement requires the shell scripts in OpenClawData/scripts/.

Automation Script Builder

Build practical automation scripts for repetitive tasks. Keep scripts simple, readable, and reliable.

Default Model

qwen2.5-coder:7b — purpose-built for scripting, file handling, and automation logic.

Storage

• Scripts → MarketingToolData/scripts/
• Logs → OpenClawData/logs/

Supported Script Types

Type	Language	Use Case
File processing	Python / Bash	Rename, move, convert, clean files
Data transformation	Python	CSV/JSON processing, data cleanup
API integration	Python	Connect to external services
Scheduling	Bash + cron	Run tasks on a schedule
Text processing	Python / Bash	Search, replace, extract, format
Report generation	Python	Compile data into reports

Script Template

Every script must include:

#!/usr/bin/env python3
"""
Script: <name>
Purpose: <one line description>
Created: YYYY-MM-DD
Usage: python3 <script-name>.py [arguments]

Dependencies: <list any pip packages needed>
"""

import sys
import os
from datetime import datetime

# Configuration
WORKSPACE_ROOT = "/Volumes/Expansion/CMO-10million"

def main():
    """Main execution logic."""
    # 1. Validate inputs
    # 2. Execute task
    # 3. Report results
    pass

if __name__ == "__main__":
    main()

Script Rules

1. Workspace guard applies — scripts must only operate within the approved workspace
2. Always include error handling — try/except with clear error messages
3. Log important actions — print what the script is doing as it runs
4. Test before deploying — run with a small test case first
5. No destructive defaults — scripts should default to safe mode (preview/dry-run)
6. Include a --dry-run flag for any script that creates, moves, or deletes files
7. Document dependencies — list any pip packages in the script header

Saving Scripts

MarketingToolData/scripts/auto-<YYYY-MM-DD>-<purpose>.py

Or for Bash:

MarketingToolData/scripts/auto-<YYYY-MM-DD>-<purpose>.sh

After Creating a Script

1. Test it with a small input
2. Show the output
3. Confirm it works before declaring done
4. Save to the scripts folder
5. Add to prompt library if it's reusable