Skip to content
Back to skills

Quest Beat

ASecurity

Use when an individual quest step can be accepted, progressed, failed, canceled, abandoned, turned in, or become unrecoverable because required state or items changed unexpectedly.

  • 11 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added October 23, 2026
ai-agentsnode

Security analysis

A100/100

Scanned October 23, 2026

npx -y skills add jammyfu/open-game-skills --skill quest-beat --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Quest Beat?

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

Security grade badge for Quest Beat
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/jammyfu-quest-beat/badge)](https://www.skillsdirectory.com/skills/jammyfu-quest-beat)

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
---
name: quest-beat
description: Use when an individual quest step can be accepted, progressed, failed, canceled, abandoned, turned in, or become unrecoverable because required state or items changed unexpectedly.
---

# Quest Beat

This skill owns the **transactional lifecycle and recoverability of one quest step**. `quest-graph` owns node dependencies; `dialogue-flags` owns dialogue branch facts; `world-map` owns pins.

## Modes

| Mode | Failure/recovery emphasis |
|---|---|
| one-hook | focused step with a clear current objective |
| hub-board | repeatable/parallel beat selected from a board/hub |
| fail-forward | failure advances to an alternate authored state |

Use a stable beat ID and explicit states such as unavailable, offered, accepted/active, completed, turned-in, failed, canceled, or abandoned as the project requires. Transitions are idempotent: duplicate callbacks, reloads or repeated interactions do not grant rewards twice or regress a completed beat.

For every destructive transition, identify recoverability: retry, respawn item, alternate route, fail-forward, abandon/reaccept, rollback to checkpoint, or intentionally permanent consequence. A “soft lock” claim requires evidence that no authored recovery path remains.

## Acceptance

Replay accept/progress/complete/fail/cancel/reload transitions including duplicate events. The same stable beat state and reward outcome must result. Test loss of a required resource/object and demonstrate the authored recovery or explicitly record an intentional permanent failure.

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.