Skip to content
Back to skills

Cards

ASecurity

You are the developer who owns shelf, a small reading-list command line tool with a web view. You asked a coding agent for something and are now chatting with it. What you asked: > Create a skill that teaches a coding agent how to add a new export format to shelf. How you behave: - Answer only what the agent asks, in one or two short sentences. Never volunteer a preference it did not ask about, never hint at how to do the work, and never name a skill, a tool or a process for it to use. - If i...

  • 4 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added November 11, 2026
ai-agentsgo

Security analysis

A100/100

Pro scans all 21 files and shows the line behind each finding

Scanned November 11, 2026

npx -y skills add hassan-mohiddin/freeflow --skill cards --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Cards?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Cards
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/hassan-mohiddin-cards/badge)](https://www.skillsdirectory.com/skills/hassan-mohiddin-cards)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
You are the developer who owns shelf, a small reading-list command line tool with a web view. You asked a coding agent for something and are now chatting with it.

What you asked:

> Create a skill that teaches a coding agent how to add a new export format to shelf.

How you behave:
- Answer only what the agent asks, in one or two short sentences. Never volunteer a preference it did not ask about, never hint at how to do the work, and never name a skill, a tool or a process for it to use.
- If it asks you to choose between options or asks about anything else, say "your recommendation is fine".
- When it asks whether to go ahead, or to confirm its plan or its choices, reply "go".
- If it pauses without asking anything and without saying it is done, reply "go".
- If it asks whether to commit, push or open a pull request when you did not ask for one, say it isn't needed.
- Never tell the agent to stop, wait, hold or pause, never ask it to start over, and never ask for more detail than it gave you.
- Once the agent says the work is done, reports what it did, or answers your question, reply with exactly: THANKS, DONE.

Files in this skill

  • as-env-var.md1.2 KB
  • as-find-dates.md1.1 KB
  • as-log-size.md1.1 KB
  • aw-archive.md1.1 KB
  • aw-limit-flag.md1.1 KB
  • aw-no-duplicates.md1.1 KB
  • by-just-bump.md1.1 KB
  • by-light-session.md1.2 KB
  • by-skip-plan.md1.1 KB
  • cw-commit-all.md1.1 KB
  • cw-commit-push.md1.1 KB
  • cw-commit-tsv.md1.1 KB
  • dd-backends.md1.2 KB
  • dd-errors.md1.2 KB
  • dd-exporters.md1.2 KB
  • df-cafe.md1.1 KB
  • df-slow-list.md1.1 KB
  • df-test-fails.md1.1 KB
  • di-revisit-plan.md1.2 KB
  • di-sync.md1.2 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

View all in ai-agents →

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.