What is a Reversibility Score (RS)?

A Reversibility Score is a metric from 0.0 to 1.0 calculated by the skill to measure how difficult, costly, or impactful it is to undo a specific decision or code change.

Can I use this skill directly in the terminal?

While it can be called, it is primarily designed as an internal 'Skill' invoked by other commands like /pr-review or /do-issue to gate complex AI-driven development tasks.

How do profiles like 'startup' or 'regulated' change the behavior?

Profiles adjust the sensitivity of the escalation trigger. The 'startup' profile allows higher-risk changes to move fast, while the 'regulated' profile triggers full deliberation for even minor architectural shifts.

Does it block my workflow if the checkpoint fails?

No, the skill includes failure handling that logs a warning and allows the calling command to continue execution to ensure the developer's momentum isn't stalled.

War Room Checkpoint

Name: War Room Checkpoint
Author: athola

byathola

•

171

•

生产力与工作流

Evaluates decision reversibility to determine if complex tasks require high-level deliberation or immediate execution.

War Room Checkpoint is a strategic decision-gate skill designed to prevent workflow bottlenecks while maintaining architectural integrity. By calculating a Reversibility Score (RS) based on factors like reversal cost, blast radius, and stakeholder impact, it intelligently decides whether a command—such as a PR review or issue implementation—can proceed automatically or needs to escalate to a full 'War Room' session. This ensures that 'Type 2' reversible decisions move fast while 'Type 1' high-stakes decisions receive the scrutiny they deserve, providing a lightweight assessment layer for embedded workflows.

主要功能

01Detailed audit logging for every decision point and assessment rationale

02Confidence-based auto-continuation to minimize unnecessary user interruptions

03171 GitHub stars

04Seamless integration with developer commands including /do-issue and /pr-review

05Automated Reversibility Score (RS) calculation across 5 critical dimensions

06Dynamic escalation logic based on customizable risk profiles like 'startup' or 'regulated'

使用场景

01Deciding if a multi-issue implementation plan should proceed automatically or pause for human review

02Determining if a PR review with multiple blocking issues needs a full architectural council

03Assessing the blast radius of breaking API changes during high-pressure sprint cycles

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add athola/claude-night-market war-room-checkpoint

For use in Claude.ai and ChatGPT

Download Skill