memory-update

Update ai-context memory files with session decisions and changes.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/fearovex/claude-config --skill memory-update
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: memory-update
Source: https://github.com/fearovex/claude-config/tree/main/skills/memory-update
Command: npx skills add https://github.com/fearovex/claude-config --skill memory-update

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Keeps the ai-context memory layer up to date by capturing and applying decisions and changes from the current work session.

Core Features & Use Cases

  • Incrementally updates ai-context files to reflect session activity (decisions, changes, and new problems).
  • Determines which files to update (stack.md, architecture.md, conventions.md, known-issues.md, changelog-ai.md, or ai-context/features/<domain>.md) based on the session.
  • Records updates without overwriting existing content, preserving history in changelog-ai.md and other memory files.

Quick Start

Run memory-update after completing a significant work session to sync session decisions into ai-context.

Frequently Asked Questions about memory-update

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

FAQPage Schema
How do I update AI context files after a development session?

To update AI context files after a session, you apply session decisions and changes incrementally to the hybrid memory layer. This captures recent activity by selectively modifying files like stack.md and architecture.md without overwriting existing content.

What is the best way to sync architecture changes to project memory non-destructively?

The best way to sync architecture changes non-destructively is through incremental updates that append new decisions to memory files. This approach preserves existing history by logging updates in changelog-ai.md rather than overwriting previous records.

When do I need to update my AI memory layer with session decisions?

You need to update your AI memory layer after completing significant work sessions, such as SDD cycles or architecture changes. Running an end-of-session review ensures the project memory accurately reflects current decisions and newly solved problems.

Which files should I update when syncing session activity into ai-context?

When syncing session activity, you update specific ai-context files based on the session's content. Targeted files include stack.md, architecture.md, conventions.md, known-issues.md, changelog-ai.md, and domain-specific feature files.

Does updating project memory overwrite existing changelog history?

No, updating project memory does not overwrite existing history. The update process uses non-destructive annotations to preserve previous entries, recording all new changes and decisions incrementally in changelog-ai.md and other memory files.