mem-weekly

Analyze weekly memory records and extract recurring behavior patterns for L2 promotion.

333|83|Updated Jan 9, 2026
One-click install
npx skills add https://github.com/SpaceZephyr/myskill --skill mem-weekly-spacezephyr
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: mem-weekly
Source: https://github.com/SpaceZephyr/myskill/tree/main/mem-weekly
Command: npx skills add https://github.com/SpaceZephyr/myskill --skill mem-weekly-spacezephyr

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps turn scattered weekly memory notes into clear, reusable behavior patterns so a personal AI memory system can identify what repeats, what matters, and what should be promoted from situation-level records into behavior-level memory.

Core Features & Use Cases

  • Weekly pattern review: Reviews the current week’s L1 situation records and identifies repeated behaviors, preferences, and tool choices.
  • Candidate generation: Automatically compiles up to five repeat-pattern candidates when the same pattern appears three or more times.
  • Human confirmation workflow: Pauses for user approval before any L2 update, allowing record, ignore, or modify-and-record decisions.
  • Memory maintenance: Updates L2 behavior files, index statistics, and weekly review logs after confirmation.
  • Use case: A user asks for a weekly review, and the Skill first scans recently modified files, then synthesizes the week’s notes into a concise set of confirmed behavior insights.

Quick Start

When the user says “周复盘”, first scan the week’s modified files, then summarize repeated patterns from the current L1 notes and wait for the user to confirm which ones should be written into L2.

Frequently Asked Questions about mem-weekly

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

FAQPage Schema
How do I extract behavior patterns from weekly notes?

To extract behavior patterns from weekly notes, this Skill scans modified L1 situation records, identifies recurring behaviors appearing three or more times, and presents candidate patterns for your confirmation before updating higher-level memory files.

What is the best way to automate weekly reviews for a personal knowledge base?

Automating weekly reviews for a personal knowledge base involves synthesizing daily logs into confirmed patterns. This Skill triggers on schedule or request, counts repeated behaviors, and compiles up to five candidate patterns for user review.

How does pattern detection work for personal memory management?

Pattern detection for personal memory management works by scanning current-month L1 notes to count repeated behaviors, preferences, and tool choices that appear at least three times, then generating candidates for promotion into a higher-level behavior layer.

Can I modify behavior patterns before they are written to L2 files?

Yes, you can modify behavior patterns before they are written to L2 files. The Skill pauses for human confirmation after generating candidates, allowing you to record, ignore, or modify-and-record decisions before any L2 update occurs.

Do I need to manually scan files to start a weekly review?

No, you do not need to manually scan files to start a weekly review. When you request a review, the Skill automatically scans recently modified files and synthesizes the week’s notes into a concise set of confirmed behavior insights.

What happens to my knowledge base if I ignore a candidate pattern?

If you ignore a candidate pattern, no changes are made to your L2 behavior files or review logs. The Skill only updates index statistics and higher-level memory layers after you explicitly approve a candidate pattern for recording.