checkpointing-an-iteration

Append recent work and decisions to the active iteration journal.

3|1|Updated Dec 12, 2025

Search Tags:#workflow #checkpoint #context-preservation #iteration-journal #AI-conversation #journal-management #state-save

Ready to run

npx skills add https://github.com/tilmon-engineering/claude-skills --skill checkpointing-an-iteration

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: checkpointing-an-iteration
Source: https://github.com/tilmon-engineering/claude-skills/tree/main/plugins/autonomy/skills/checkpointing-an-iteration
Command: npx skills add https://github.com/tilmon-engineering/claude-skills --skill checkpointing-an-iteration

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill enables saving current iteration progress by updating the journal entry to preserve state before potential context loss.

Core Features & Use Cases

Incremental progress saving: appends recent work, decisions, artifacts, and context to the active iteration journal.
Context preservation before risky actions: ensures important information survives compaction and resets.
Repeatable checkpoint workflow: provides a deterministic process to locate the current journal, review progress, and commit updates.

Quick Start

Run the /checkpoint-iteration command to save progress. After running, locate the journal at autonomy/[goal-name]/iteration-NNNN-YYYY-MM-DD.md and verify the "Work Performed" section contains the latest entries.

Frequently Asked Questions about checkpointing-an-iteration

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

FAQPage Schema

How do I save AI conversation state before context compaction?▼

To save AI conversation state before context compaction, use a checkpointing workflow that appends recent work, decisions, and artifacts to an existing iteration journal, preserving chronological order and ensuring important information survives potential context loss.

What is the best way to preserve context during long AI conversation iterations?▼

The best way to preserve context during long AI conversation iterations is to execute a deterministic checkpoint workflow that updates your iteration journal entry with incremental progress before pausing or taking risky actions.

How do I create a checkpoint to save progress in an iteration journal?▼

To create a checkpoint and save progress in an iteration journal, run the checkpoint command to append new information to the active journal file located at autonomy/[goal-name]/iteration-NNNN-YYYY-MM-DD.md.

When should I save iteration progress to prevent context loss?▼

You should save iteration progress to prevent context loss during long conversations, before mid-iteration pauses, or immediately prior to executing risky actions where important context must be explicitly preserved.

Do I need an existing journal file to save conversation checkpoints?▼

Yes, you need an existing journal file to save conversation checkpoints because the checkpoint workflow relies on appending new information to a pre-existing journal while maintaining chronological order.

How do I verify my iteration journal checkpoint was saved correctly?▼

To verify your iteration journal checkpoint was saved correctly, locate the journal file at autonomy/[goal-name]/iteration-NNNN-YYYY-MM-DD.md and confirm the "Work Performed" section contains your latest entries.