memory

Query a memory store for prior solutions before scanning a codebase.

1|Updated Nov 12, 2025
One-click install
npx skills add https://github.com/grahama1970/fetcher --skill memory-grahama1970
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: memory
Source: https://github.com/grahama1970/fetcher/tree/main/.agents/skills/memory
Command: npx skills add https://github.com/grahama1970/fetcher --skill memory-grahama1970

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Memory-first problem solving reduces redundant codebase exploration by querying stored knowledge before diving into code.

Core Features & Use Cases

  • Recall prior solutions before examining the repository to accelerate debugging.
  • Learn new insights to enrich future recalls and build a knowledge graph.
  • Support end-to-end workflow: recall -> scan (only if needed) -> learn.

Quick Start

Start memory workflow by running recall to check for a prior solution before scanning the codebase.

Frequently Asked Questions about memory

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

FAQPage Schema
How do I reduce redundant codebase exploration when debugging recurring issues?

Reduce redundant codebase exploration by querying a memory store to recall prior solutions before scanning the repository. This memory-first approach accelerates debugging by retrieving stored knowledge and only examining code if no previous answer exists.

What is the memory-first workflow for troubleshooting a codebase?

The memory-first workflow follows a recall, scan, and learn sequence. You first query stored knowledge for prior solutions, scan the codebase only if needed, and then learn new insights to enrich future recalls and build a knowledge graph.

Do I need a memory service to recall prior solutions before scanning code?

Yes, you need a memory service with recall and learn endpoints, alongside a local memory client. These dependencies are required to store and retrieve past answers, enabling the automated problem-solving workflow before codebase exploration.

How do I capture new debugging insights to build a knowledge graph?

Capture new debugging insights by using the learn endpoint after scanning the codebase. This enriches future recalls by storing newly discovered solutions in the memory client, building a knowledge graph for subsequent troubleshooting workflows.

When should I avoid memory-first codebase recall and just scan the repository?

Avoid memory-first codebase recall when troubleshooting entirely novel issues lacking prior solutions, or when the memory service and local memory client are unavailable. In these cases, proceed directly with codebase scanning to debug.