continuation-recovery

Search episodic memory to recover missing prior-session decisions and preferences.

Updated Feb 7, 2026
One-click install
npx skills add https://github.com/thisolivier/chronolog --skill continuation-recovery
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: continuation-recovery
Source: https://github.com/thisolivier/chronolog/tree/main/.claude/skills/continuation-recovery
Command: npx skills add https://github.com/thisolivier/chronolog --skill continuation-recovery

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It prevents you from losing important decisions, gotchas, and preferences when a conversation continues but prior context is missing or not visible.

Core Features & Use Cases

  • Auto-triggered session recovery: Detects continuation cues (summary markers, “continue” intent, or references to completed work) and initiates targeted recovery.
  • Episodic memory search for just-in-time recall: Searches episodic memory using the project name, branch, and task keywords to retrieve relevant prior outcomes.
  • Actionable context extraction and minimal memory updates: Pulls only decisions, gotchas, and preferences, then appends a compact recovery section to MEMORY.md if needed.

Quick Start

Say continue where we left off so the skill can recover the missing decisions, gotchas, and preferences and then resume your task.

Frequently Asked Questions about continuation-recovery

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

FAQPage Schema
How do I recover missing context when continuing a previous session?

Recover missing context by triggering episodic memory search using project name, branch, and task keywords to retrieve prior decisions, gotchas, and preferences not visible in the current continuation. The skill extracts actionable items, updates MEMORY.md if needed, and provides a brief handoff before proceeding.

Why does my conversation continuation lose prior decisions and preferences?

Conversations lose prior decisions and preferences when the continuation context lacks visible references to completed work or summary markers. Session continuation recovery detects these cues, searches episodic memory for just-in-time recall, and rebuilds missing context without wasting tokens on irrelevant history.

How do I resume a task and pick up where we left off without wasting tokens?

To pick up where you left off without wasting tokens, say "continue where we left off" to auto-trigger episodic memory search. The skill retrieves only actionable decisions, gotchas, and preferences, appends a compact recovery section to MEMORY.md if needed, and proceeds with the task.

What is episodic memory search for session handoffs?

Episodic memory search for session handoffs is a just-in-time recall mechanism that queries prior outcomes using scoped project, branch, and task keywords. It extracts only actionable items like decisions, gotchas, and preferences to rebuild missing context efficiently during continuation scenarios.

When should I avoid using automated context recovery for session continuation?

Avoid automated context recovery when the current conversation already contains the necessary visible context, or when prior decisions, gotchas, and preferences are already captured in MEMORY.md. The skill targets only continuation scenarios referencing completed work not present in the current context window.