memory-access

Read and write global agent memory files via terminal shell scripts.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/zcawood-mie/agentz --skill memory-access
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: memory-access
Source: https://github.com/zcawood-mie/agentz/tree/main/skills/memory-access
Command: npx skills add https://github.com/zcawood-mie/agentz --skill memory-access

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Provides a way for agents to read and write global memory files outside the workspace without triggering VS Code external-file approval prompts or leaving visible editor changes, enabling persistent cross-project context and caches.

Core Features & Use Cases

  • Read global memory files located under the agents root and stream their contents to the agent without creating editor-visible files.
  • Write complete memory files atomically, creating parent directories as needed and replacing the full file contents to avoid incremental edits.
  • Use cases include storing stable user configuration, skill caches like a PR dashboard, and cross-project session state that must persist across workspaces.

Quick Start

Use the memory-access skill to read or write a global memory file such as pr-dashboard using the provided terminal scripts.

Frequently Asked Questions about memory-access

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

FAQPage Schema
How do I store persistent agent memory across different workspaces?▼

You can store persistent agent memory across workspaces by using terminal-invoked shell scripts to read and write files under the global ~/.agents/memories directory. This approach keeps cross-project caches outside the active workspace.

How do I write global memory files atomically from terminal scripts?▼

Writing global memory files atomically is done through full-file replacements via terminal shell scripts. The script creates parent directories as needed and replaces the entire file content to avoid incremental edits.

Can I access cross-project caches without triggering VS Code external-file approval prompts?▼

Yes, you can access cross-project caches without triggering VS Code external-file approval prompts by operating on memory files located under the agents root directory via terminal scripts rather than the editor interface.

What happens when a terminal script tries to read a missing memory file?▼

When a terminal script attempts to read a missing memory file, it returns a clear exit code indicating the file is missing. The scripts provide distinct exit codes for missing files, usage errors, and success.

What is the best way to manage cross-project session state for agents?▼

The best way to manage cross-project session state is using terminal-invoked scripts to read and write global memory files. This ensures session state persists across workspaces without leaving visible editor changes.

Why does my agent need prompt-free access to global memory files?▼

Prompt-free access to global memory files is needed to store stable user configuration and caches seamlessly. It allows agents to operate on persistent context outside the workspace without manual approval interruptions.