harnesscp-memory

Design and audit a file-based memory system for VS Code Copilot.

Updated Mar 31, 2026
One-click install
npx skills add https://github.com/LSDIPPOLLC/harness-cp --skill harnesscp-memory
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: harnesscp-memory
Source: https://github.com/LSDIPPOLLC/harness-cp/tree/main/.github/skills/harnesscp-memory
Command: npx skills add https://github.com/LSDIPPOLLC/harness-cp --skill harnesscp-memory

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill solves the problem of Copilot forgetting important user preferences, project decisions, and external references between sessions by creating a durable file-based memory system.

Core Features & Use Cases

  • Memory System Design: Plan and structure .github/memory/ with an index and typed memory files.
  • Memory Auditing: Check existing memories for staleness, duplication, missing frontmatter, and broken index links.
  • Persistent Context Recovery: Support cross-session recall for user roles, feedback, project status, and references when the assistant keeps losing context.
  • Use Case: When a team says “I already told you this,” use this Skill to capture the correction as persistent memory and keep future sessions aligned.

Quick Start

Use the harnesscp-memory skill to design, audit, or repair a Copilot memory system for a project that needs persistent context across conversations.

Frequently Asked Questions about harnesscp-memory

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

FAQPage Schema
How do I give VS Code Copilot persistent memory across sessions?▼

Give VS Code Copilot persistent memory by designing a file-based system with a .github/memory index, typed memory files with frontmatter, and SessionStart/Stop hooks for cross-session recall of user preferences and project decisions.

Why does Copilot keep forgetting my project decisions and user preferences?▼

Copilot forgets project decisions because it lacks a durable memory system. You can fix this context loss by applying a structured file-based memory index with typed files and curator-agent workflows to store and recover data across conversations.

How do I audit Copilot memory files for staleness and broken index links?▼

Audit Copilot memory files by checking existing memories in the .github/memory directory for staleness, duplication, missing frontmatter, and broken index links to ensure reliable cross-session recall and context recovery.

Can I use VS Code hooks to recover persistent context for Copilot?▼

Yes, you can use VS Code hooks to recover persistent context. SessionStart and Stop hooks integrate with a .github/memory index and typed memory files to trigger reliable cross-session recall of project status and references.

What's the best way to structure project memory for an AI coding assistant?▼

The best way to structure project memory for an AI coding assistant is to create a .github/memory directory containing an index and typed memory files with frontmatter, managed by curator-agent workflows for reliable recall.

Do I need a .github/memory index to maintain Copilot state across conversations?▼

Yes, a .github/memory index is required to maintain Copilot state. It organizes typed memory files with frontmatter and supports curator-agent workflows that ensure reliable cross-session recall of user roles and feedback.