memory-query

Name: memory-query
Availability: InStock
Author: Valynt

Read, write, and query agent memory across in-process and persistent backend layers.

Updated Nov 17, 2025

Search Tags:#memory #query #cross-session #semantic-memory #memory-system #tenant-isolation #vector-memory

Ready to run

npx skills add https://github.com/Valynt/ValueOS --skill memory-query-valynt

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: memory-query
Source: https://github.com/Valynt/ValueOS/tree/main/.windsurf/skills/memory-query
Command: npx skills add https://github.com/Valynt/ValueOS --skill memory-query-valynt

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Memory Query skill enables agents to read, write, and search across in-process MemorySystem and persistent memory in packages/memory, ensuring tenant isolation and cross-session recall.

Core Features & Use Cases

Read/write memory inside a single execution using MemorySystem for fast, deterministic access.
Cross-session memory recall, vector search, and provenance with packages/memory for longer-term memory.
Frontend restrictions: memory data remains backend-only to protect data integrity.

Quick Start

Retrieve prior hypotheses from memory and store a new semantic memory entry for downstream agents.

Frequently Asked Questions about memory-query

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

FAQPage Schema

How do I retrieve and store agent memory across different sessions?▼

Cross-session memory recall is managed through a persistent memory package that enables vector search and provenance tracking. It stores semantic and episodic memory entries so downstream agents can retrieve prior hypotheses and stakeholder data across backend layers.

What is the difference between in-process memory and persistent semantic memory?▼

In-process memory provides fast, deterministic read and write access inside a single execution, while persistent semantic memory stores data for long-term cross-session recall. Both operate under layered access with tenant isolation enforced across backend layers.

How do I query semantic memory while maintaining tenant isolation?▼

You query semantic memory through a layered access system that enforces tenant isolation across all backend operations. This restricts memory data to authorized tenants during cross-session read, write, and vector search operations.

Can I access cross-session vector memory directly from the frontend?▼

No, cross-session vector memory data remains backend-only to protect data integrity. Frontend restrictions are explicitly enforced to prevent direct access, ensuring all memory queries route securely through backend layers.

What's the best way to search prior hypotheses stored in an agent memory system?▼

The best way to search prior hypotheses is using the persistent memory package's vector search and provenance capabilities. This allows you to retrieve semantic memory entries stored by downstream agents across multiple sessions.

Does this memory query approach work for managing episodic memory in backend layers?▼

Yes, this approach manages semantic and episodic memory across backend layers using a layered access design. It handles read, write, and query operations while enforcing tenant isolation for secure cross-session recall.