checkpoint-trigger

Natural language wrapper for checkpoint commands - automatically triggers /checkpoint:create, /checkpoint:restore, /checkpoint:list when users request checkpoint operations

⭐5stars🔀1forks📁View on GitHub🕐Updated Jan 11, 2026

When & Why to Use This Skill

The checkpoint-trigger skill provides a seamless natural language interface for managing development states through automated checkpoint commands. By translating conversational intent into git-based operations like create, restore, and list, it empowers developers to maintain a reversible and safe coding workflow without manual command execution.

Use Cases

  • Risk-Free Refactoring: Save the current project state using natural language before starting a complex code overhaul, allowing for an immediate rollback if errors occur.
  • Experimental Development: Quickly 'save state' before testing new libraries or experimental logic, ensuring a clean recovery point is always available.
  • Version History Management: Use simple phrases like 'show checkpoints' to view a list of saved states and navigate through different iterations of a feature during the implementation phase.
  • State Recovery: Effortlessly restore a specific working version by asking the agent to 'load checkpoint' when a debugging session leads to a dead end.
namecheckpoint-trigger
descriptionNatural language wrapper for checkpoint commands - automatically triggers /checkpoint:create, /checkpoint:restore, /checkpoint:list when users request checkpoint operations
schema_version1.0

checkpoint-trigger

Type: WRITE-CAPABLE DAIC Modes: IMPLEMENT only Priority: Low

Trigger Reference

This skill activates on:

  • Keywords: "create checkpoint", "save checkpoint", "restore checkpoint", "load checkpoint", "list checkpoints", "checkpoint", "save state", "restore state"
  • Intent Patterns: (create|save|make).*?checkpoint, (restore|load|recover).*?checkpoint, (list|show).*?checkpoint, (save|restore).*?state

From: skill-rules.json - checkpoint-trigger configuration

Purpose

Automatically trigger checkpoint commands (/checkpoint:create, /checkpoint:restore, /checkpoint:list) when users request checkpoint operations using natural language.

Core Behavior

When activated in IMPLEMENT mode:

  1. Checkpoint Intent Detection

    • Detect checkpoint operations from natural language
    • Route to appropriate checkpoint command
  2. Command Routing

    • Create: "save checkpoint" → /checkpoint:create
    • Restore: "load checkpoint X" → /checkpoint:restore X
    • List: "show checkpoints" → /checkpoint:list

Natural Language Examples

Triggers this skill:

  • ✓ "Create checkpoint"
  • ✓ "Save current state"
  • ✓ "Restore checkpoint 3"
  • ✓ "List all checkpoints"

Safety Guardrails

WRITE-CAPABLE RULES:

  • ✓ Only execute in IMPLEMENT mode
  • ✓ Checkpoints use git stash mechanism
  • ✓ Verify clean working directory before restore