agent-framework-redis

What it is and what it does

agent-framework-redis integrates Redis as a persistent backend for Microsoft Agent Framework agents. It provides two main components: RedisContextProvider for storing and retrieving agent context and user preferences across sessions, and RedisHistoryProvider for maintaining conversation history in Redis Lists with automatic message trimming and thread isolation.

The package is designed for production use with connection pooling, error handling, and performance optimization built in. It enables distributed agent deployments where multiple instances can share conversation state, and allows chat history to survive application restarts. You supply a Redis connection URL (supporting local Docker, Redis Cloud, or Azure Managed Redis) and use the providers as you would any other Agent Framework memory backend.

Use it for:

• Build multi-turn chatbots where agents remember user preferences and conversation context across separate chat sessions
• Deploy distributed agent systems where multiple application instances share conversation history and context state
• Implement user session management with automatic cleanup of old messages while preserving recent conversation history
• Create agents that survive application restarts without losing conversation state or learned user preferences
• Support thread-isolated conversations in concurrent agent deployments using Redis key partitioning

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Adds Redis-backed persistent memory and conversation history to Microsoft Agent Framework agents, enabling them to remember context and chat history across sessions and application restarts.

Yes, if you are building agents with Microsoft Agent Framework and need persistent memory or conversation history. The package is actively maintained, has no known vulnerabilities, low install friction, and permissive licensing. The only prerequisite is a running Redis instance, which the documentation covers clearly. Not necessary if you only need in-memory or file-based storage.

Install

agent-framework-redis on PyPI

pip install agent-framework-redis

uv add agent-framework-redis

poetry add agent-framework-redis

Installing agent-framework-redis

Before you install

Low friction install with a pure-Python wheel. Actively maintained as of 2026-08-14 with 12804 repository stars. Depends on agent-framework-core, redis, redisvl, and numpy—all standard packages with no compiled dependencies beyond what numpy requires.

License in practice

Licensed under MIT (permissive), so you can use, modify, and distribute this package freely in commercial and private projects without restriction.

Quickstart

pip install agent-framework-redis --pre

from agent_framework_redis import RedisContextProvider, RedisHistoryProvider

# Requires a running Redis instance (local, Docker, or cloud)
context_provider = RedisContextProvider(redis_url="redis://localhost:6379")
history_provider = RedisHistoryProvider(redis_url="redis://localhost:6379")

Requires a running Redis server (local, Docker, or cloud). See the package description for setup options including Docker (redis:8.0.3), Redis Cloud, or Azure Managed Redis.

Verify before relying

• Whether RedisContextProvider and RedisHistoryProvider are the only memory/history providers available in this package
• Performance characteristics under high message volume or concurrent thread access
• Exact serialization format and compatibility guarantees across Agent Framework versions

Package facts

Evidence: agent_framework_redis-1.0.0b260813-py3-none-any.whl

Tags

More Application Frameworks packages

Further reading