Skip to content

Feedback on your automating-database-backups skill #293

@RichardHightower

Description

@RichardHightower

I took a look at your automating-database-backups skill and wanted to share some thoughts.

Links:

The TL;DR

You're at 54/100 — solidly in failing territory. The good news? You nailed Spec Compliance (14/15) — your YAML structure and naming conventions are clean. The rough part is Utility (8/20) and PDA (14/30) — the skill promises functionality that doesn't actually exist yet, and the structure needs better organization to make this easier to use.

What's Working Well

  • Strong trigger coverage — "automate database backups", "schedule database dumps", "disaster recovery" — these are the exact terms developers search for
  • Multi-database support — PostgreSQL, MySQL, MongoDB, SQLite is a solid range
  • Proper metadata structure — Your YAML frontmatter is valid and follows conventions perfectly

The Big One: Stub Scripts Without Real Functionality

Your scripts directory has 4 Python files (backup_script_generator.py, backup_validator.py, etc.), but they're all template stubs. backup_script_generator.py literally says "Add processing logic here" — it's generic file processing, not actual backup generation.

Why this matters: Developers will load your skill expecting to get working backup scripts. Instead they get templates. This tanks your Utility score hard.

The fix: Implement at least one working script. Start with PostgreSQL since it's common:

  • backup_script_generator.py should actually generate valid pg_dump commands with compression, rotation, and encryption options
  • backup_validator.py should have real restore testing logic
  • Keep the stubs for other databases as examples of the pattern

This alone could add +5 points.

Other Things Worth Fixing

  1. Missing reference files — Your README lists 6 files (postgresql_backup_restore.md, etc.) marked unchecked. Either create them or remove the list. Right now it reads like broken promises. Move your verbose error handling and database-specific details into actual reference files instead of cramming 140 lines into SKILL.md. (+6 points)

  2. Non-existent template paths — You reference {baseDir}/templates/backup-scripts/postgresql-backup.sh in Resources, but these don't exist. Pick one: create the files or remove the references. Don't leave developers hanging. (+4 points)

  3. Empty Examples section — It just says "Example usage patterns will be demonstrated in context." Add 2-3 real examples: scheduling a daily PostgreSQL backup to S3, backing up MongoDB with encryption, restoring from a backup timestamp. (+3 points)

  4. Bloated SKILL.md — 140 lines with 30 sub-steps is overwhelming. Trim it to ~50 lines with the essentials, move step details and error handling to references. Better navigation for developers = better UX. (+5 points)

Quick Wins

  • Fix the description grammar: change "Process use when" to "Use when automating database backups"
  • Create the 6 reference files from your README (or remove the list)
  • Implement one working backup script (PostgreSQL is the obvious choice)
  • Add concrete examples to the Examples section
  • Move verbose error handling out of SKILL.md into a separate reference file

These changes get you from 54 → 70+ pretty quickly. Focus on making at least one backup script actually functional — that's your biggest bang for buck.


Checkout your skill here: [SkillzWave.ai](https://skillzwave.ai) | [SpillWave](https://spillwave.com) We have an agentic skill installer that install skills in 14+ coding agent platforms. Check out this guide on how to improve your agentic skills.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions