Agent-Skills.md

Agent Skills: Python Scripter

Create robust Python automation with full logging and safety checks. Use when tasks need complex data processing, authenticated API work, conditional file operations, or error handling beyond simple shell commands.

UncategorizedID: vaayne/cc-plugins/python-script

Repository

vaayne/cc-plugins
vaayneLicense: MIT
261

Install this agent skill to your local

pnpm dlx add-skill https://github.com/vaayne/agent-kit/tree/HEAD/skills/python-script

Skill Files

Browse the full folder contents for python-script.

Download Skill

Loading file tree…

skills/python-script/SKILL.md

Skill Metadata

Name
python-script
Description
Create robust Python automation with full logging and safety checks. Use when tasks need complex data processing, authenticated API work, conditional file operations, or error handling beyond simple shell commands.

Python Scripter

Overview

Design and run Python scripts with clear requirements, safety checks, and reproducible logging.

Workflow

{project} is the git repo name (if in a git repo) or the basename of the current working directory.

  1. Confirm inputs, outputs, constraints, and preferred libraries.
  2. Identify risky operations and secure explicit approval.
  3. Scaffold a script with PEP 723 metadata and structured logging.
  4. Lint with uvx ruff check --fix ~/.agents/sessions/{project}/scripts/{script_name}.py.
  5. Run with uv run --script ~/.agents/sessions/{project}/scripts/{script_name}.py and monitor logs.
  6. Report results, risks encountered, and any follow-up steps.

Logging requirements

  • Log to ~/.agents/sessions/{project}/logs/{script_name}.log and stream to console
  • Capture start/end timestamps, parameters, file operations, and errors
  • Add RotatingFileHandler if logs may grow large

Template

Read references/script-template.md for the full script template.

Safety practices

  • Provide dry-run or rollback paths
  • Validate external inputs and API responses
  • Use temporary directories for intermediates
  • Never run destructive steps without confirmation