overnight-mode

Automate unattended overnight development with checkpointing and safety constraints.

Updated Jan 16, 2026
One-click install
npx skills add https://github.com/adiomas/claude-code-adiomas-plugins --skill overnight-mode
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: overnight-mode
Source: https://github.com/adiomas/claude-code-adiomas-plugins/tree/main/autonomous-dev/skills/overnight-mode
Command: npx skills add https://github.com/adiomas/claude-code-adiomas-plugins --skill overnight-mode

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill provides the governance, context, and runtime rules to allow Claude to operate in unattended overnight development without user interaction, while preserving safety and traceability.

Core Features & Use Cases

  • NO USER INTERACTION: runs autonomously with clear decision logging and checkpoint updates.
  • STRICT TDD and iteration tracking: enforces a red-green-refactor cycle for overnight tasks.
  • STATE & CHECKPOINT MANAGEMENT: maintains active/started state, deadlines, and progress in .claude files.
  • VERIFICATION & REPORTING: logs issues, tests, and git history to support post-run review.
  • Auto-restart on context limits: gracefully checkpoints and resumes when limits approach.

Quick Start

Enable overnight mode by configuring the initial state files and starting the automation loop.

Frequently Asked Questions about overnight-mode

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run autonomous development tasks overnight without user interaction?

Unattended overnight development requires governance rules, explicit state files, and checkpointing to automate autonomous development tasks safely without user interaction. You configure initial state files and start the automation loop to enable this mode.

What is checkpointing and how does it manage state during unattended code generation?

Checkpointing during unattended code generation maintains active state, deadlines, and progress tracking in explicit state files. It ensures traceability by logging decisions and supports auto-restart on context limits by gracefully saving and resuming operations.

Can I enforce strict TDD cycles for automated overnight development tasks?

Yes, automated overnight development tasks can enforce strict TDD cycles. The automation enforces a red-green-refactor testing discipline, tracking iterations to ensure code quality and preventing unverified changes from being committed during unattended operation.

How do I review autonomous coding results and audit logs after an overnight run?

Reviewing autonomous coding results after an overnight run involves checking structured reporting outputs. The system logs issues, test results, and git history to support post-run review, providing full traceability of the automated development session.

Does unattended overnight development support auto-restart when context limits are reached?

Unattended overnight development supports auto-restart when context limits are approached. It gracefully checkpoints the current state and resumes the automation loop, ensuring long-running tasks complete without manual intervention or data loss.

What are the limitations of running non-interactive autonomous development sessions?

Limitations of non-interactive autonomous development sessions include the strict requirement for predefined scope and safety constraints. Without user interaction, the system relies entirely on initial configuration, state files, and strict rules to prevent unhandled errors or unsafe operations.