What are atomic writes in this context?

Atomic writes ensure that a file is either fully updated or not updated at all by writing to a temporary file first and then performing a rename operation, preventing partial data corruption if a crash occurs.

How do I implement file locking for concurrent access?

The skill provides a specialized context manager pattern using system-level locking to acquire exclusive access to state files, preventing race conditions when multiple processes attempt to write simultaneously.

How does the crash recovery pattern work?

It uses progress tracking and checkpoints to allow an agent to identify exactly where it left off, enabling the system to skip completed tasks and resume only pending work after an interruption.

Is this skill compatible with Claude Code 2.0?

Yes, it is specifically designed for Claude Code 2.0+, utilizing a progressive disclosure architecture to optimize token usage and context efficiency.

State Management Patterns

Name: State Management Patterns
Author: akaszubski

byakaszubski

•

数据库管理

Implements robust state persistence and crash recovery patterns for autonomous software agents.

关于

This skill provides a standardized framework for managing state within the Claude Code ecosystem, focusing on reliability and data integrity. It offers production-ready implementation patterns for atomic JSON writes, file locking for concurrency, and comprehensive crash recovery mechanisms. By using this skill, developers can ensure that their autonomous agents maintain context across restarts, handle simultaneous file access without corruption, and gracefully migrate state schemas as applications evolve through versioning and migration logic.

主要功能

Automated crash recovery with progress tracking and checkpoints
Exclusive file locking for safe concurrent state access and modification
Schema versioning and migration paths for seamless state evolution
10 GitHub stars
Atomic JSON writes to prevent file corruption during system crashes
Progressive disclosure architecture for optimized token consumption

使用场景

Managing user preferences and session tracking across multiple Claude restarts
Implementing safe concurrent data access in multi-agent environments
Building long-running batch processing agents that need to resume after failure

关于

主要功能

Automated crash recovery with progress tracking and checkpoints
Exclusive file locking for safe concurrent state access and modification
Schema versioning and migration paths for seamless state evolution
10 GitHub stars
Atomic JSON writes to prevent file corruption during system crashes
Progressive disclosure architecture for optimized token consumption

使用场景

Managing user preferences and session tracking across multiple Claude restarts
Implementing safe concurrent data access in multi-agent environments
Building long-running batch processing agents that need to resume after failure