skills.rest
winsorllc
winsorllcCOMMUNITY
Explore all skills by this author →
GitHub

memory-search

Store, search, and retrieve memories across jobs using JSON files.

1|Updated Feb 25, 2026
Search Tags:#context management#knowledge base#memory#persistence#long-term memory#agent learning
One-click install
npx skills add https://github.com/winsorllc/upgraded-carnival --skill memory-search-winsorllc
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: memory-search
Source: https://github.com/winsorllc/upgraded-carnival/tree/main/.pi/skills/memory-search
Command: npx skills add https://github.com/winsorllc/upgraded-carnival --skill memory-search-winsorllc

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires natural, node-fetch, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill provides a persistent memory for agents, allowing them to recall information, learn from past interactions, and build a knowledge base across multiple jobs.

Core Features & Use Cases

  • Long-term Context: Store and retrieve information that persists between agent sessions.
  • Knowledge Base Building: Create a structured repository of learned information.
  • Learning from Experience: Agents can reference past decisions and outcomes to improve future performance.
  • Use Case: An agent can store user preferences from one job and recall them in a subsequent job to personalize its responses.

Quick Start

Use the memory store command to save the fact that the user prefers dark mode.

Frequently Asked Questions about memory-search

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

FAQPage Schema
How do I add long-term memory to an agent so it remembers context across jobs?▼

You establish long-term memory by using a file-based index and individual JSON files to store, search, and retrieve agent memories across multiple jobs. This enables persistent context and knowledge base construction between sessions.

What is the best way to build an agent knowledge base from past interactions?▼

Building an agent knowledge base involves storing past decisions and outcomes as individual JSON files with tags and categories. Agents can then search this repository to learn from experience and improve future performance.

How do I store and retrieve user preferences for agent personalization?▼

You store and retrieve user preferences by saving them as memories using dedicated commands. The agent can then recall these specific memories in subsequent jobs to personalize its responses based on past interactions.

Can I filter agent memories by tags, categories, and time?▼

Yes, you can filter agent memories by tagging, categorization, and time-based filtering. This allows efficient memory management and retrieval when searching through the file-based index of stored JSON files.

What are the limitations of using a file-based index for agent memory persistence?▼

Using a file-based index for agent memory persistence relies on individual JSON files, which may face scalability constraints compared to database solutions. It is designed for storing, searching, and retrieving context rather than high-throughput data processing.

Related Skills

Explore More Software Engineering Skills→
TbK0ngTbK0ng

trellis-bootstrap

Initialize Trellis AI workflow scaffolding with .trellis structure, memory, and CLI hooks.

Community
Intermediate
FVossebeldFVossebeld

context-optimization

Reduce token usage by masking, compacting, caching, and partitioning context.

Community
Advanced
0xUrsanomics0xUrsanomics
2

dreaming

Ingest session logs, detect patterns, and promote scored memories into project trees.

Community
Advanced
zerwizzerwiz

pinecone

Manage Pinecone vector database workflows for AI retrieval applications.

Community
Intermediate