skills.rest
jwilger
jwilgerCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

memory-protocol

Store and retrieve agent knowledge in markdown files with grep-based search.

Updated Jul 1, 2026
Search Tags:#recall#memory#grep#file-based#knowledge-graph#agent-memory
One-click install
npx skills add https://github.com/jwilger/claude-code-plugins --skill memory-protocol
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: memory-protocol
Source: https://github.com/jwilger/claude-code-plugins/tree/main/sdlc/skills/memory-protocol
Command: npx skills add https://github.com/jwilger/claude-code-plugins --skill memory-protocol

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a structured approach to accumulating and retrieving institutional knowledge across Claude Code agent tasks, reducing repeated work and context loss.

Core Features & Use Cases

  • Recall-before-act workflow: search memory before starting tasks or debugging.
  • Remember-after-discovery: store insights, conventions, and decisions for future reuse.
  • Knowledge graph connectivity: link related memories to enable richer context and discovery.

Quick Start

  • After solving a problem, store a concise memory entry under the Memory Directory using a simple markdown format.
  • Before starting a new task, recall related memories using grep to surface past patterns.
  • Periodically review and connect memories to build a lightweight knowledge graph.

Frequently Asked Questions about memory-protocol

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

FAQPage Schema
How do I build durable agent memory with fast recall in Claude Code?▼

Build durable agent memory using a file-based store of markdown entries. Recall related memories with grep search before starting tasks to surface past patterns and reduce repeated work across workflows.

What is the best way to store debugging insights for future agent tasks?▼

Store debugging insights as concise markdown entries in a memory directory after solving a problem. This remember-after-discovery approach ensures past conventions and decisions inform current work.

How does grep-based search work for retrieving agent knowledge?▼

Grep-based search retrieves agent knowledge by scanning markdown files in the memory directory for keywords. This lightweight approach surfaces relevant memories quickly without needing a database.

Can I create a knowledge graph from file-based memories without a database?▼

Yes, create a knowledge graph from file-based memories by adding simple links between related markdown entries. Periodically reviewing and connecting these memories enables richer context and navigable discovery.

Does this file-based memory system require any external dependencies?▼

No external dependencies are required. The system relies on lightweight markdown files and grep-based search, making it easy to integrate into Claude Code workflows without complex setup.

When should I use a recall-before-act workflow for task planning?▼

Use a recall-before-act workflow before starting any task planning, debugging, or design decisions. Searching accumulated memories first prevents context loss and leverages institutional knowledge for current work.

Related Skills

Explore More Software Engineering Skills→
danielcopperdanielcopper
3

memory-dream

Triage daily markdown logs to promote recurring themes into durable knowledge bases.

Community
Intermediate
ScottRBKScottRBK
3

exploring-knowledge-graph

Traverse Forgetful knowledge graphs to map memories, entities, and relationships.

Community
Advanced
xoaixoai
25

memory

Integrate sage-memory into Sage workflows for persistent knowledge management.

Community
Advanced
karrtik159karrtik159

rag-implementation

Build RAG systems with vector databases and semantic search.

Community
Advanced