Skip to content
Back to skills

Plan Queue

ASecurity

Use when players pause or slow execution to author future commands and the project needs stable plan/command identity, validation, ordering and commit semantics without inventing a second simulation clock.

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

Security analysis

A100/100

Scanned October 23, 2026

npx -y skills add jammyfu/open-game-skills --skill plan-queue --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Plan Queue?

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

Security grade badge for Plan Queue
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/jammyfu-plan-queue/badge)](https://www.skillsdirectory.com/skills/jammyfu-plan-queue)

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: plan-queue
description: Use when players pause or slow execution to author future commands and the project needs stable plan/command identity, validation, ordering and commit semantics without inventing a second simulation clock.
---

# Plan queue

Ask the planning mode: freeze-queue, soft-pause, live-plan, none, or existing.

## Contract

Publish:
- stable `plan_session_id` and plan revision
- stable `command_id` for each queued command
- actor/target/action references and authored parameters
- validation time and invalidation policy
- explicit `commit order` / concurrency semantics
- cancel/edit/confirm request identity
- execution owner and pause policy reference

## Ownership

`plan-queue` owns authored command intent and commit ordering. `pause-timescale` owns time policy. The action/combat/navigation system that receives a command owns whether and how it executes.

## Rules

1. Planning does not create another combat/simulation clock.
2. Commands may execute simultaneously, sequentially, dependency-ordered or actor-authored according to policy; same-tick execution is not universal.
3. Confirm/edit/cancel requests are idempotent by plan/request identity.
4. A command invalidated between planning and execution records a deterministic skip/replan/fallback result rather than silently retargeting.
5. Presentation order in the queue UI is not execution order unless explicitly authored.

## Acceptance

Given the same `plan_session_id`, revision, queued commands and confirm request, the same command set and commit order are produced. A reviewer can distinguish plan validity from the downstream systems that execute each action.

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.