feat(ai): add AI libraries directory and metadata for the integrations page

README.md

@@ -54,10 +54,10 @@ lines of three dashes (`---`). These are YAML fields of which we use only the
 `title` field (and possibly `linkTitle`). The title field is used instead of an
 H1 heading in each of the pages.
 
-### Clients, modules, libraries, tools
+### Clients, modules, libraries, tools, and AI libraries
 
-We maintain links to clients, modules, libraries and tools in various languages in
-JSON files stored under `clients/`, `modules/`, `libraries/` and `tools/`
+We maintain links to clients, modules, libraries, tools, and AI libraries in various languages in
+JSON files stored under `clients/`, `modules/`, `libraries/`, `tools/`, and `ai/`
 respectively.
 
 **Note**:  Clients listed here, while fully compatible with Valkey, are not all official clients for Valkey.
@@ -81,7 +81,7 @@ For example [clients/go/valkey-go.json](./clients/go/valkey-go.json):
 }
 ```
 
-Modules, libraries and tools follow a similar structure under their respective directories.
+Modules, libraries, tools and AI libraries follow a similar structure under their respective directories.
 
 ### Commands
 

ai/README.md

@@ -0,0 +1,57 @@
+# AI Libraries
+
+This directory contains JSON files for AI libraries that are compatible with Valkey.
+
+## Structure
+
+AI libraries are organized by language:
+
+```
+ai/
+  python/
+    library-name.json
+  javascript/
+    library-name.json
+  ...
+```
+
+## JSON Format
+
+Each AI library JSON file should contain the following fields:
+
+```json
+{
+    "name": "Library Name",
+    "description": "One or two sentence description of the AI library and why someone would use it",
+    "valkey_uses": "Describes how Valkey is used (e.g., vector store, semantic search, AI memory, caching)",
+    "repo": "https://github.com/org/repo",
+    "installation": "pip install library-name",
+    "version": "1.0.0",
+    "version_released": "2025-01-15",
+    "language": "python",
+    "license": "MIT"
+}
+```
+
+### Field Descriptions
+
+- **name** (string, required): The name of the AI library
+- **description** (string, required): A concise 1-2 sentence description of the library and why someone would use it
+- **valkey_uses** (string, required): Explanation of how Valkey is used within the library (vector store/search, semantic search, AI memory, caching, session storage, etc.)
+- **repo** (string, required): URL to the library's source code repository
+- **installation** (string, required): Installation command (e.g., `pip install`, `npm install`)
+- **version** (string, required): Current version number (semantic versioning)
+- **version_released** (string, required): Date when the current version was released (YYYY-MM-DD format)
+- **language** (string, required): Primary programming language of the library
+- **license** (string, required): Software license type
+
+## Adding a New AI Library
+
+1. Create a subdirectory for the language if it doesn't exist (e.g., `python/`, `javascript/`)
+2. Create a JSON file named after the library (lowercase, hyphens for spaces)
+3. Fill in all required fields
+4. Submit a pull request
+
+## Examples
+
+See the existing JSON files in this directory for reference implementations.

ai/java/langchain4j.json

@@ -0,0 +1,11 @@
+{
+    "name": "Langchain4j",
+    "description": "Idiomatic Java library for building LLM applications with unified API over LLM providers, supporting RAG, tool calling, and enterprise Java frameworks.",
+    "valkey_uses": "Embedding store for vector search using Valkey's vector search capabilities (Valkey 8+)",
+    "repo": "https://github.com/langchain4j/langchain4j-community",
+    "installation": "Add dependency: dev.langchain4j:langchain4j-community-valkey:1.15.0-beta25",
+    "version": "1.15.0-beta25",
+    "version_released": "2026-05-15",
+    "language": "java",
+    "license": "Apache-2.0"
+}

ai/javascript/openmemory.json

@@ -0,0 +1,11 @@
+{
+    "name": "OpenMemory JavaScript SDK",
+    "description": "Create self-hosted cognitive memory for AI agents with multi-sector memory (episodic, semantic, procedural, emotional) and explainable recall traces.",
+    "valkey_uses": "Memory storage backend for cognitive memory sectors with temporal knowledge graph support and decay-based retention",
+    "repo": "https://github.com/CaviraOSS/OpenMemory",
+    "installation": "npm install openmemory-js",
+    "version": "1.2.2",
+    "version_released": "2025-12-06",
+    "language": "javascript",
+    "license": "Apache-2.0"
+}

ai/javascript/recall.json

@@ -0,0 +1,11 @@
+{
+    "name": "Recall",
+    "description": "MCP server providing persistent cross-session memory for Claude and AI coding agents, surviving context compaction with automatic hooks.",
+    "valkey_uses": "Memory storage backend for facts, decisions, and conversation history across AI agent sessions",
+    "repo": "https://github.com/joseairosa/recall",
+    "installation": "npx @modelcontextprotocol/create-server @joseairosa/recall",
+    "version": "1.7.0",
+    "version_released": "2025-12-02",
+    "language": "javascript",
+    "license": "MIT"
+}

ai/lua/kong.json

@@ -0,0 +1,11 @@
+{
+    "name": "Kong Gateway",
+    "description": "Cloud-native API gateway with powerful plugins for traffic control, authentication, and observability for microservices and AI APIs.",
+    "valkey_uses": "Backend for rate limiting counters, caching responses, and session storage in distributed API gateway deployments",
+    "repo": "https://github.com/Kong/kong",
+    "installation": "docker pull kong/kong-gateway:3.14",
+    "version": "3.14",
+    "version_released": "2025-11-20",
+    "language": "lua",
+    "license": "Apache-2.0"
+}

ai/python/cocoindex.json

@@ -0,0 +1,11 @@
+{
+    "name": "CocoIndex",
+    "description": "Incremental data indexing framework for AI agents with real-time updates, supporting codebases, documents, and knowledge bases with minimal reprocessing.",
+    "valkey_uses": "Vector store for semantic code search and document embeddings with incremental indexing support",
+    "repo": "https://github.com/cocoindex-io/cocoindex",
+    "installation": "pip install cocoindex",
+    "version": "1.0.12",
+    "version_released": "2026-06-20",
+    "language": "python",
+    "license": "Apache-2.0"
+}

ai/python/cognee.json

@@ -0,0 +1,11 @@
+{
+    "name": "Cognee Valkey Adapter",
+    "description": "Build AI agent memory using Cognee's Extract-Cognify-Load (ECL) pipelines to replace RAG with structured, queryable knowledge graphs.",
+    "valkey_uses": "Vector database adapter for storing and retrieving embeddings in AI memory systems with semantic search",
+    "repo": "https://github.com/topoteretes/cognee-community",
+    "installation": "pip install cognee-community-vector-adapter-valkey",
+    "version": "0.4.0",
+    "version_released": "2025-12-28",
+    "language": "python",
+    "license": "Apache-2.0"
+}

ai/python/db-gpt.json

@@ -0,0 +1,11 @@
+{
+    "name": "DB-GPT",
+    "description": "AI-native data application framework with Agentic Workflow Expression Language (AWEL) for building database-aware agents and workflows.",
+    "valkey_uses": "Vector store for semantic similarity search and cache store for LLM response caching in database query optimization",
+    "repo": "https://github.com/eosphoros-ai/DB-GPT",
+    "installation": "pip install dbgpt",
+    "version": "0.8.1",
+    "version_released": "2026-06-17",
+    "language": "python",
+    "license": "MIT"
+}

ai/python/haystack.json

@@ -0,0 +1,11 @@
+{
+    "name": "Haystack Valkey Integration",
+    "description": "Build production-ready retrieval-augmented generation (RAG) pipelines using Haystack's modular components for NLP and LLM applications.",
+    "valkey_uses": "Document store with vector similarity search for semantic retrieval and document embeddings in RAG pipelines",
+    "repo": "https://github.com/deepset-ai/haystack-core-integrations",
+    "installation": "pip install valkey-haystack",
+    "version": "0.3.0",
+    "version_released": "2026-07-22",
+    "language": "python",
+    "license": "Apache-2.0"
+}

ai/python/kserve.json

@@ -0,0 +1,11 @@
+{
+    "name": "KServe",
+    "description": "Kubernetes-native model serving platform for deploying and scaling ML models with autoscaling, networking, and health checking built-in.",
+    "valkey_uses": "Indirect LMCache remote KV-cache storage",
+    "repo": "https://github.com/kserve/kserve",
+    "installation": "pip install kserve",
+    "version": "0.17.0",
+    "version_released": "2026-03-13",
+    "language": "python",
+    "license": "Apache-2.0"
+}

ai/python/langchain-aws.json

@@ -0,0 +1,11 @@
+{
+    "name": "LangChain AWS",
+    "description": "Build LLM applications on AWS infrastructure with LangChain integrations for Bedrock, SageMaker, and other AWS services.",
+    "valkey_uses": "Cache store for LLM responses and vector store for semantic search with AWS ElastiCache for Valkey or MemoryDB",
+    "repo": "https://github.com/langchain-ai/langchain-aws",
+    "installation": "pip install langchain-aws",
+    "version": "1.4.1",
+    "version_released": "2026-03-19",
+    "language": "python",
+    "license": "MIT"
+}

ai/python/open-webui.json

@@ -0,0 +1,11 @@
+{
+    "name": "Open-WebUI",
+    "description": "Self-hosted ChatGPT-like interface for running local LLMs with Ollama and OpenAI-compatible APIs, supporting RAG and offline operation.",
+    "valkey_uses": "Session storage and caching for chat histories, user preferences, and RAG document embeddings",
+    "repo": "https://github.com/open-webui/open-webui",
+    "installation": "pip install open-webui",
+    "version": "0.9.6",
+    "version_released": "2026-06-01",
+    "language": "python",
+    "license": "Open WebUI (modified BSD-3)"
+}

ai/python/openmemory.json

@@ -0,0 +1,11 @@
+{
+    "name": "OpenMemory Python SDK",
+    "description": "Create self-hosted cognitive memory for AI agents with multi-sector memory (episodic, semantic, procedural, emotional) and explainable recall traces.",
+    "valkey_uses": "Memory storage backend for cognitive memory sectors with temporal knowledge graph support and decay-based retention",
+    "repo": "https://github.com/CaviraOSS/OpenMemory",
+    "installation": "npm i openmemory-js",
+    "version": "1.3.0",
+    "version_released": "2025-12-06",
+    "language": "typescript",
+    "license": "Apache-2.0"
+}

ai/python/praisonai.json

@@ -0,0 +1,11 @@
+{
+    "name": "PraisonAI",
+    "description": "Low-code framework for creating and orchestrating multi-agent AI systems with AutoGen and CrewAI support for autonomous task execution.",
+    "valkey_uses": "Caching layer for LLM responses and session management to optimize multi-agent workflows and reduce API costs",
+    "repo": "https://github.com/MervinPraison/PraisonAI",
+    "installation": "pip install praisonai",
+    "version": "4.6.46",
+    "version_released": "2026-05-21",
+    "language": "python",
+    "license": "MIT"
+}

ai/python/vllm.json

@@ -0,0 +1,11 @@
+{
+    "name": "vLLM",
+    "description": "Fast LLM inference and serving engine optimized for throughput with PagedAttention, supporting production deployments with high performance.",
+    "valkey_uses": "Caching backend for KV cache management, vector storage for semantic routing, and agentic memory components",
+    "repo": "https://github.com/vllm-project/vllm",
+    "installation": "pip install vllm",
+    "version": "0.15.1",
+    "version_released": "2026-02-04",
+    "language": "python",
+    "license": "Apache-2.0"
+}

resources/ai/index.md

@@ -0,0 +1,10 @@
+---
+title: "AI Libraries"
+linkTitle: "AI Libraries"
+weight: 12
+description: AI and machine learning libraries that support Valkey for caching, vector storage, and data management
+layout: bazzar
+bazzar: ai
+aliases:
+    - /resources/ai/
+---