skills.rest
Agency7ai
Agency7aiOFFICIAL
Explore all skills by this author →
WebsiteGitHub

rag-memory

Store and retrieve project memories across sessions using a Convex backend.

Updated Dec 24, 2025
Search Tags:#context#convex#vector-search#memory-management#retrieval-augmented-generation#rag-memory
One-click install
npx skills add https://github.com/Agency7ai/agent-rag-memories-universal --skill rag-memory
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: rag-memory
Source: https://github.com/Agency7ai/agent-rag-memories-universal/tree/main
Command: npx skills add https://github.com/Agency7ai/agent-rag-memories-universal --skill rag-memory

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

RAG Memory provides a portable, framework-agnostic solution to preserve and retrieve project context across Claude Code or agent sessions by storing memories in a Convex backend and injecting relevant history into conversations.

Core Features & Use Cases

  • Framework-agnostic persistence: store decisions, patterns, progress, blockers, and context so you can continue work across sessions.
  • Semantic load and context augmentation: load memories by topic or pull recent items to inform responses.
  • Easy setup and collaboration: switch between local backends or shared Convex deployments; use with any tech stack.

Quick Start

Copy this skill into your project, run /rag-memory setup, then start using /rag-memory load and /rag-memory save to manage context.

Frequently Asked Questions about rag-memory

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

FAQPage Schema
How do I preserve project context across Claude Code sessions?▼

To preserve project context across sessions, you can use retrieval-augmented generation to store decisions, code patterns, and progress in a Convex backend, then load relevant memories to inform your next conversation.

What is the best way to retrieve past agent decisions and blockers?▼

Retrieving past agent decisions and blockers is best handled through semantic memory loads, allowing you to pull specific context types like progress or file summaries from a vector-search database to augment your current session.

How do I set up persistent memory for retrieval-augmented generation?▼

To set up persistent memory for retrieval-augmented generation, copy the skill into your project, run the setup command to configure the Convex backend, then use save and load commands to manage context.

Can I use RAG memory management with any tech stack or framework?▼

Yes, RAG memory management is framework-agnostic and requires no framework code in client applications, allowing you to integrate persistent context retrieval into any tech stack seamlessly.

Does context augmentation support shared team deployments?▼

Context augmentation supports both local backends and shared Convex deployments, enabling collaboration by allowing multiple sessions to access the same stored project memories and retrieved context.

What types of memories can I store for session context augmentation?▼

You can store several memory types for session context augmentation, including decision, code_pattern, progress, blocker, context, and file_summary, enabling structured retrieval of project history.

Related Skills

Explore More Software Engineering Skills→
mamd69mamd69
4

Ruvector Setup

Configure Ruvector vector storage with REDB persistence and HNSW indexing for Node.js projects.

Community
Intermediate
firefly-eventsfirefly-events

test

Coordinate multiple agents to author, execute, and triage end-to-end tests.

Official
Advanced
SoftSystemsStudioSoftSystemsStudio

context-session

Manages Claude Code session context via CLI save/load commands.

Community
Intermediate
agentuityagentuity
1

agentuity-cli-cloud-vector-upsert

Upsert vectors with metadata and embeddings into a cloud vector store.

Official
Advanced