What problem does it solve?
This Skill ensures a coding agent’s important working context survives disconnects and interruptions, so the work can be resumed with minimal loss rather than starting over.
Core Features & Use Cases
- Durable session directory: Creates a timestamped
session/ workspace and keeps checkpoints organized for a long-running or multi-step effort.
- Human-readable state & resume artifacts: Maintains
session_state.md, timeline.md, files.md, and handoff.md so another agent can quickly understand goals, progress, and next actions.
- Checkpoint rhythm & recovery: Establishes when to write checkpoints and how to recover by reading
handoff.md first, then session_state.md, then recent timeline entries.
Use case: You start implementing a risky refactor and later you must disconnect; upon returning, another agent reads the latest handoff and timeline to continue from the verified next action.
Quick Start
Ask the AI to “start a new session memory checkpoint for this work” and let it create session/<timestamp>/ with session_state.md, timeline.md, files.md, and handoff.md.