skills.rest
rockywuest
rockywuestCOMMUNITY
Explore all skills by this author →
𝕏@rockywuestGitHub

auto-checkpoint

Inject the last operational state from a checkpoint file into agent sessions.

3|1|Updated Mar 8, 2026
Search Tags:#state management#checkpointing#agent memory#context continuity#session persistence#data backup
One-click install
npx skills add https://github.com/rockywuest/openclaw-memory-local --skill auto-checkpoint
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: auto-checkpoint
Source: https://github.com/rockywuest/openclaw-memory-local/tree/main/auto-checkpoint
Command: npx skills add https://github.com/rockywuest/openclaw-memory-local --skill auto-checkpoint

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill ensures your AI agent maintains its operational state and context across sessions, preventing it from "forgetting" its progress or important information.

Core Features & Use Cases

  • State Injection: Automatically injects the last known operational state into the agent's context at the start of each new session.
  • Staleness Detection: Warns you if the checkpoint is older than a configurable threshold, prompting you to update it.
  • Compaction Backup: Creates a timestamped backup of the checkpoint file before the agent's context is compacted, safeguarding against data loss.
  • Use Case: For an agent managing a long-term project, this ensures it remembers all previous tasks, decisions, and ongoing discussions every time you interact with it, even after restarts or context cleanups.

Quick Start

Instruct the agent to use the auto-checkpoint skill to inject the last operational state into the current session.

Frequently Asked Questions about auto-checkpoint

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

FAQPage Schema
How do I maintain agent memory and context continuity across multiple sessions?▼

To maintain context continuity, you inject a checkpoint file containing the last operational state into every new agent session. This ensures the agent always knows where it left off without losing previous progress.

Why does my AI agent forget previous tasks after a context compaction or restart?▼

Your agent forgets tasks after a context compaction because operational state is not persisted. Creating a timestamped backup of your checkpoint file before compaction safeguards against this data loss.

How do I automatically inject operational state into an agent at the start of a session?▼

You can automatically inject operational state by reading a checkpoint file like state/current.md and passing it as context. Instruct the agent to use the auto-checkpoint skill to load this state.

Can I get a warning when my agent's checkpoint or session persistence file is stale?▼

Yes, staleness detection warns you if the checkpoint file is older than a configurable threshold. This prompts you to update the operational state before continuing the session.

What is the best way to back up agent memory before context compaction?▼

The best way to back up agent memory is creating a timestamped backup of the checkpoint file before context compaction occurs. This prevents data loss and preserves the session persistence state.

Related Skills

Explore More Software Engineering Skills→
alansmelloalansmello

speckit-git-initialize

Initialize a Git repository with an initial commit in Bash or PowerShell.

Community
Intermediate
furrycolombia-sysfurrycolombia-sys
1

checkpoint

Create a checkpoint file recording git status, pending tasks, and context.

Community
Intermediate
hbvg234hbvg234
1

pre-compact-backup

Back up Claude transcripts before context compaction with JSON metadata.

Community
Intermediate
DektoraDektora

rotation-handoff

Generate and read structured DekSpec session handoff records in dekspec/.scratch/rotation-handoff/.

Official
Advanced