iris-development

Promote transient session events into durable, searchable long-term memories.

94|22|Updated Jan 29, 2026
One-click install
npx skills add https://github.com/redis/agent-skills --skill iris-development
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: iris-development
Source: https://github.com/redis/agent-skills/tree/main/skills/iris-development
Command: npx skills add https://github.com/redis/agent-skills --skill iris-development

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Iris provides a unified interface to manage Redis Agent Memory (RAM) across session memory and long-term memory, enabling AI agents to persist and semantically search memory with a background promotion workflow.

Core Features & Use Cases

  • Two-tier memory model: session memory for raw, ordered events and long-term memory for semantically searchable facts.
  • Background promotion worker to extract durable facts and populate long-term memory from sessions.
  • Official SDKs for Python and TypeScript to create, authenticate, and query memory stores, plus clear guidance on prompts, tokens, and TTLs for memory lifecycle.
  • Use cases include recording conversations, retrieving context for new conversations, and performing semantic searches across memories in Redis Cloud.

Quick Start

Provision an Iris RAM store in Redis Cloud, install the Python or TypeScript SDK, configure server URL and API key, then begin appending session events and querying long-term memories.

Frequently Asked Questions about iris-development

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

FAQPage Schema
How do I persist AI agent session memory into searchable long-term storage?

To persist AI agent memory, you append transient session events to a memory store and use a background promotion worker to extract durable facts into long-term memory for semantic search in Redis Cloud.

What is the difference between session memory and long-term memory for AI agents?

Session memory stores raw, ordered conversational events for immediate context, while long-term memory holds semantically searchable facts promoted from sessions to provide durable context for new AI agent conversations.

How do I configure TTLs for Redis AI agent memory stores?

You configure memory lifecycle TTLs during the initial store setup in Redis Cloud, defining separate expiration durations for both transient session memory and promoted long-term memory events.

Can I use TypeScript or Python to manage Redis agent memory?

Yes, official Python and TypeScript SDKs allow you to create, authenticate, and query memory stores, enabling you to append session events and perform semantic searches across long-term memories.

Do I need Redis Cloud to use the Iris RAM service for AI memory orchestration?

Yes, the Iris RAM service requires a configured store provisioned within Redis Cloud, along with a server URL and bearer token authentication to manage and query your AI agent memory.

Why does my AI agent lose context across new conversation sessions?

Context is lost when sessions are transient. A background promotion worker extracts durable facts from session memory into long-term memory, allowing semantic retrieval of past context for new conversations.