MCP/Redis MCP Server

Redis MCP Server

Official Model Context Protocol server for Redis. Manage keys, data structures, streams, pub/sub, JSON documents, and vector search.

rag-data•Databases & Storage

GitHub640

License:MIT

VERIFIED

@Redis

About this Server

The official Redis Model Context Protocol (MCP) server provides a natural language interface for agentic workflows to inspect, cache, manage, and search data across Redis. It features complete support for strings, hashes, lists, sets, sorted sets, streams consumer groups, pub/sub channels, JSON document manipulation, vector search indexes, and natural language documentation search.

Real-time Interaction

Direct low-latency RPC protocol communication with structured JSON schemas.

Scoped Sandboxing

Restricted capability bounds to prevent accidental modifications or leakage.

Security & Best Practices

TLS Encryption: Use rediss:// scheme with ssl_ca_certs validation when connecting to remote cloud instances.

Credential Handling: Pass passwords and connection strings via environment variables or secret managers rather than command history.

Database Number Scoping: Scope agent operations to a specific logical database index (e.g. /0 or /1) to isolate critical data.

When to Use (and When Not To)

Best For

•High-speed ephemeral caching, session state management, and TTL-based record expiration
•Real-time event streams, activity feeds, task queues, and pub/sub message broadcasting
•Hybrid in-memory vector similarity searches and nested JSON document retrieval

Not For / Limitations

•Complex relational multi-table SQL queries and foreign key joins (use DBHub, Neon, or PlanetScale MCP instead)
•Heavy OLAP analytical aggregations over large Parquet datasets (use DuckDB / MotherDuck MCP instead)
•Document collections needing complex MongoDB aggregation pipelines (use MongoDB MCP instead)

Install & Setup

One-line terminal command

terminal command

$claude mcp add redis -- uvx --from redis-mcp-server@latest redis-mcp-server --url redis://localhost:6379/0

Exposed Tools & Endpoints

10 tools

Functions your AI agent can invoke via this Model Context Protocol server

string_get_set• callable tool

Sets and retrieves string keys with optional TTL expiration for caching and session data.

hash_operations• callable tool

Stores and reads individual field-value pairs and vector embeddings inside Redis hashes.

list_operations• callable tool

Appends, pops, and inspects list items for message queues and FIFO/LIFO workloads.

set_sorted_set• callable tool

Manages unique sets and score-ordered sorted sets for leaderboards and priority queues.

pubsub_publish_subscribe• callable tool

Publishes messages and manages channel/pattern subscriptions for real-time notifications.

streams_consumer_groups• callable tool

Reads, acknowledges, and manages stream event sourcing and consumer groups.

json_get_set• callable tool

Stores, queries, and mutates nested JSON documents using JSONPath expressions.

vector_search• callable tool

Manages vector indexes and executes similarity searches over stored embeddings.

redis_docs_search• callable tool

Searches official Redis documentation, tutorials, and command reference manuals.

server_info• callable tool

Retrieves Redis server memory stats, active clients, and connection information.