memory-merger

Merge mature lessons from domain memory files into instruction files.

Updated Apr 14, 2026
One-click install
npx skills add https://github.com/BrunoAMSilva/my-config --skill memory-merger-brunoamsilva
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: memory-merger
Source: https://github.com/BrunoAMSilva/my-config/tree/main/coding/skills/memory-merger
Command: npx skills add https://github.com/BrunoAMSilva/my-config --skill memory-merger-brunoamsilva

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Consolidates mature learnings from scattered domain memory files into their canonical instruction files to preserve important knowledge, reduce redundancy, and make guidance more scannable and maintainable.

Core Features & Use Cases

  • Propose-for-review: Reads domain-memory and instruction files, extracts candidate memories, and presents a structured list for human approval before any file changes.
  • Iterative merging: Drafts merged instructions, evaluates against a defined quality bar (zero knowledge loss, minimal redundancy, maximum scannability), and refines until approved.
  • Safe updates: Creates or updates instruction files, merges applyTo patterns comprehensively, and removes merged sections from the memory file, supporting both global (user) and workspace scopes.
  • Use Case: A developer consolidates prompt-engineering tips collected in a memory file into a polished, reusable prompt instruction that applies across projects.

Quick Start

Run the memory-merger command with a domain and optional scope, for example /memory-merger >prompt-engineering to propose and merge global prompt memories into the instruction file.

Frequently Asked Questions about memory-merger

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

FAQPage Schema
How do I consolidate scattered domain memory files into instruction files in VS Code?

To consolidate domain memory files, this tool extracts mature lessons and proposes them for human review before merging them into canonical instruction markdown files. It supports both global and workspace scopes to preserve knowledge and reduce redundancy.

What is the best way to merge applyTo patterns and clean up memory entries during knowledge maintenance?

Merging applyTo patterns and cleaning memory entries is handled by iteratively drafting and validating merged instructions against a quality bar. Once approved, it updates the instruction file and removes the merged sections from the memory file.

How does propose-for-review work when merging memories into instructions?

The propose-for-review mechanism reads domain-memory and instruction files, extracts candidate memories, and presents a structured list for human approval before any file changes occur. This ensures zero knowledge loss and minimal redundancy during the merge.

Can I use this to merge workspace-specific prompt-engineering tips into a global instruction file?

Yes, you can merge workspace-specific or global prompt-engineering memories into a polished, reusable instruction file. It operates on VS Code instruction directories to comprehensively merge applyTo patterns across different project scopes.

Do I need to manually validate merged instructions against a quality bar?

Manual validation is not required because the tool iteratively drafts and evaluates merged instructions against a defined quality bar of zero knowledge loss, minimal redundancy, and maximum scannability. It refines the instructions internally until they meet these criteria.

What happens to memory entries after they are merged into an instruction file?

After memories are merged into the instruction file, the tool automatically removes the merged sections from the source memory file. This cleaning process prevents redundancy and keeps your knowledge base maintainable.