Redis Agent Skills - Setup Guide¶
Source: redis/agent-skills (163⭐, MIT)
Skill family: redis/agent-skills (12 indexed skills)
Combined Installs: ~19,866 across indexed listings (Oct 1, 2026 snapshot)
Category: Data / Infrastructure
Quality Tier: 🟡 Trusted (official Redis, Inc. publisher; MIT LICENSE in-repo; low star count for an official org)
Redis ships an official collection of agent skills covering the operational decisions that agents most often get wrong: choosing the right data structure, client/connection strategy, security posture, and observability. The skills are authored by Redis, Inc. (author: Redis, Inc. in each SKILL.md) and carry per-skill MIT licenses. The cluster includes AI-focused guidance (iris-development, redis-semantic-cache) alongside classic data-modeling and cluster-operation skills.
Installation¶
npx skills add redis/agent-skills
Roster - Indexed Skills¶
| Skill | Installs | What It Does |
|---|---|---|
| redis-core | 3,337 | Data-structure modeling - String, Hash, List, Set, Sorted Set, JSON, Stream, Vector Set |
| redis-development | 3,258 | Umbrella plugin entry for the Redis development skill set |
| redis-connections | 2,375 | Connection pooling, multiplexing, pipelining, RESP3 client-side caching |
| redis-security | 2,055 | auth (requirepass/ACL users), TLS, least-privilege ACL, network exposure |
| redis-observability | 1,967 | Metrics to monitor, built-in commands, diagnosing memory/connection issues |
| redis-semantic-cache | 1,645 | Redis LangCache - semantic caching of LLM responses on Redis Cloud |
| redis-clustering | 1,583 | Cluster/replication, hash tags, avoiding CROSSSLOT, reading from replicas |
| iris-development | 1,451 | Iris (Redis AI umbrella) - Agent Memory (RAM) data plane on Redis Cloud |
| redis-search | 1,308 | FT.CREATE schema design, field types, DIALECT 2 queries, vector search |
| redis-query-engine | 453 | Query engine guidance |
| redis-vector-search | 433 | Vector search guidance |
| redis-best-practices | 1 | Best-practices stub (minimal installs) |
CorpusIQ Use Cases¶
| Use Case | How |
|---|---|
| Cache design | redis-semantic-cache + redis-core to plan an LLM response cache and key model |
| Agent memory | iris-development for Redis Agent Memory as an agent's long-term store |
| Production hardening | redis-security + redis-observability before shipping a Redis-backed service |
| Search/vector | redis-search + redis-vector-search for hybrid RAG retrieval on Redis |
Integration with CorpusIQ¶
CorpusIQ is the connector layer that lets AI assistants read the systems a business already runs on - accounting, payments, CRM, email, files and analytics - and answer with the source named. It is read-only by design: it retrieves data and never writes to, or acts in, a connected system, which is what makes it safe to point at live accounts.
This page is part of the CorpusIQ directory, the reference we maintain for the MCP server and agent-skill ecosystem. Run alongside CorpusIQ, a third-party server like this one can be combined with a company's own business data in a single answer.
Limitations / Verification¶
- skills.sh indexing verified Oct 1, 2026: 12 skills indexed; 8 SKILL.md files verified via raw.githubusercontent on branch
main. - Repo has 16 SKILL.md files: each skill is duplicated under
skills/andplugins/redis-development/skills/(plugin layout). Not all skills are indexed on skills.sh. - Star count (163) is low for an official org repo - the "Trusted" tier reflects the publisher identity (Redis, Inc.), not community traction. Verify before relying in production.
- No live install test performed; install counts are from the Oct 1, 2026 sweep snapshot.
Security¶
| Audit | Verdict |
|---|---|
| Gen Agent Trust Hub | Not published |
| Socket | Not published |
| Snyk | Not published |
Related¶
← Skills Catalog | Marketplace → Powered by CorpusIQ