reflect

Analyze the current session and consolidate learnings. Use at the end of a session or task.

⭐88stars🔀9forks📁View on GitHub🕐Updated Dec 30, 2025

When & Why to Use This Skill

The 'reflect' skill is a specialized Claude agent tool designed to enhance session continuity and knowledge retention. It systematically analyzes the current workspace context, extracts key decisions and completed tasks, and archives them into a persistent history. This process ensures that critical project insights are preserved and that the agent's active memory remains clutter-free for subsequent tasks, significantly improving long-term productivity and context awareness.

Use Cases

  • Project Session Handoff: Use at the end of a complex task to generate a structured summary of progress, ensuring the next session starts with a clear understanding of what was achieved and what remains.
  • Decision Logging: Automatically capture and archive the rationale behind technical or strategic decisions made during a session to build a searchable historical record.
  • Context Window Optimization: Prevent performance degradation by clearing the 'current' scratchpad and moving relevant data to long-term storage, keeping the agent's immediate focus sharp and efficient.
  • Automated Progress Reporting: Generate concise updates on tasks completed and open issues to be used for status reports or team synchronization.
namereflect
descriptionAnalyze the current session and consolidate learnings. Use at the end of a session or task.

Reflect

Consolidate the current session's work into the recall history.

Workflow

  1. Read Context: Read .fleet/context/recall/current.md to see what happened this session.

  2. Summarize: Create a brief summary of:

    • Tasks completed
    • Key decisions made
    • Open issues
  3. Archive: Append this summary to .fleet/context/recall/history.md (or creating a dated file in recall/sessions/).

  4. Clear Scratchpad: Reset .fleet/context/recall/current.md for the next task.