aget-save-state

Save workflow state as ISO 8601 timestamped YAML checkpoint files.

1|Updated Oct 4, 2025
One-click install
npx skills add https://github.com/gmelli/public-OpenAI-DeepResearch-aget --skill aget-save-state-gmelli
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: aget-save-state
Source: https://github.com/gmelli/public-OpenAI-DeepResearch-aget/tree/main/.claude/skills/aget-save-state
Command: npx skills add https://github.com/gmelli/public-OpenAI-DeepResearch-aget --skill aget-save-state-gmelli

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Save workflow state to prevent context loss and enable reliable resume and recovery for long-running or interruptible sessions. This Skill externalizes session context so humans and agents can pick up work mid-flow without reestablishing state.

Core Features & Use Cases

  • Automatic and Named Checkpoints: Create timestamped checkpoints automatically or use sanitized, human-friendly names for identifiable resumes.
  • Structured YAML State Capture: Persist session metadata including session file path, concise current focus, active tasks, pending questions, modified artifacts, uncommitted git status, and a human-readable resume prompt.
  • Safety and Governance: Enforce inviolable constraints such as forbidden directories, exclusion of sensitive data, ISO 8601 timestamps, directory creation for sessions/checkpoints, and overwrite semantics.
  • Use Case: Save a checkpoint before a major refactor, prior to context-intensive research calls, or at natural breakpoints to enable later recovery and continuity.

Quick Start

Save the current session as checkpoint design-refactor to capture context, active tasks, modified files, and a recommended next action for later resumption.

Frequently Asked Questions about aget-save-state

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

FAQPage Schema
How do I save workflow state for session continuity and recovery?

To save workflow state, you create timestamped YAML checkpoint files that capture your session file path, current focus, active tasks, and artifacts. This enables reliable resume and recovery for long-running or interruptible sessions without reestablishing context.

When do I need to create session checkpoints for long-running research workflows?

You need session checkpoints before context-intensive operations, major refactors, or anticipated session interruptions. Creating checkpoints at natural breakpoints externalizes your current focus and active tasks, preventing context loss mid-flow.

How do I resume a CLI-based session after an interruption using a checkpoint?

To resume a session, you load a previously generated checkpoint file containing the session file path, active tasks, and a human-readable resume prompt. This structured YAML state capture allows agents and humans to pick up work instantly.

Does the checkpoint YAML capture include sensitive credentials or uncommitted git status?

The checkpoint YAML capture explicitly excludes sensitive credentials to enforce safety governance. However, it does include uncommitted git status, modified artifacts, and pending questions to provide complete workflow context for recovery.

What is the best way to structure session state persistence for automated recovery?

The best way to structure state persistence is using ISO 8601 timestamped YAML files stored in a dedicated sessions/checkpoints directory. This approach enforces sanitization of names and overwrite semantics for identifiable, reliable resumes.

Can I use named checkpoints instead of automatic timestamps for session management?

Yes, you can use sanitized, human-friendly names to create identifiable checkpoints instead of relying solely on automatic timestamps. This allows for easier identification when you need to resume work from a specific project phase.