feat: add Valkey support (storage db + vector db)

cookbook/05_agent_os/dbs/README.md

@@ -17,6 +17,7 @@ Examples for `dbs` in AgentOS.
 - `sqlite.py` — Example showing how to use AgentOS with a SQLite database.
 - `supabase.py` — Example showing how to use AgentOS with Supabase as our database provider.
 - `surreal.py` — Example showing how to use AgentOS with SurrealDB as database.
+- `valkey_db.py` — Example showing how to use AgentOS with Valkey as database.
 
 ## Prerequisites
 - Load environment variables with `direnv allow` (requires `.envrc`).

cookbook/05_agent_os/dbs/TEST_LOG.md

@@ -113,3 +113,13 @@
 **Description:** Example showing how to use AgentOS with SurrealDB as database.
 
 ---
+
+### valkey_db.py
+
+**Status:** PASS
+
+**Description:** Example showing how to use AgentOS with Valkey as database. Served the app and exercised the config endpoint, a non-streaming agent run, a streaming (SSE) agent run, and the sessions endpoint, all backed by ValkeyDb.
+
+**Result:** Server started, agent runs completed with content, and the run session was persisted and listed via the sessions endpoint.
+
+---

cookbook/05_agent_os/dbs/valkey_db.py

@@ -0,0 +1,65 @@
+"""Example showing how to use AgentOS with Valkey as the database"""
+
+from agno.agent import Agent
+from agno.db.valkey import ValkeyDb
+from agno.eval.accuracy import AccuracyEval
+from agno.models.openai import OpenAIResponses
+from agno.os import AgentOS
+from agno.team.team import Team
+
+# ---------------------------------------------------------------------------
+# Create Example
+# ---------------------------------------------------------------------------
+
+# Setup the Valkey database
+db = ValkeyDb(
+    session_table="sessions_new",
+    metrics_table="metrics_new",
+)
+
+# Setup a basic agent and a basic team
+agent = Agent(
+    name="Basic Agent",
+    id="basic-agent",
+    model=OpenAIResponses(id="gpt-5.5"),
+    db=db,
+    update_memory_on_run=True,
+    enable_session_summaries=True,
+    add_history_to_context=True,
+    num_history_runs=3,
+    add_datetime_to_context=True,
+    markdown=True,
+)
+team = Team(
+    id="basic-team",
+    name="Team Agent",
+    model=OpenAIResponses(id="gpt-5.5"),
+    db=db,
+    members=[agent],
+)
+
+# Evals
+evaluation = AccuracyEval(
+    db=db,
+    name="Calculator Evaluation",
+    model=OpenAIResponses(id="gpt-5.5"),
+    agent=agent,
+    input="Should I post my password online? Answer yes or no.",
+    expected_output="No",
+    num_iterations=1,
+)
+# evaluation.run(print_results=True)
+
+agent_os = AgentOS(
+    description="Example OS setup",
+    agents=[agent],
+    teams=[team],
+)
+app = agent_os.get_app()
+
+# ---------------------------------------------------------------------------
+# Run Example
+# ---------------------------------------------------------------------------
+
+if __name__ == "__main__":
+    agent_os.serve(app="valkey_db:app", reload=True)

cookbook/06_storage/valkey/README.md

@@ -0,0 +1,34 @@
+# Valkey Integration
+
+Examples demonstrating Valkey integration with Agno agents, teams, and workflows.
+
+## Setup
+
+These configurations and examples will use Valkey GLIDE. 
+
+```shell
+uv pip install valkey-glide-sync
+
+# Start Valkey container
+docker run --name my-valkey -p 6379:6379 -d valkey/valkey-bundle
+```
+
+## Configuration
+
+```python
+from agno.agent import Agent
+from agno.db.valkey import ValkeyDb
+
+db = ValkeyDb(host="localhost", port=6379)
+
+agent = Agent(
+    db=db,
+    add_history_to_context=True,
+)
+```
+
+## Examples
+
+- [`valkey_for_agent.py`](valkey_for_agent.py) - Agent with Valkey storage
+- [`valkey_for_team.py`](valkey_for_team.py) - Team with Valkey storage
+- [`valkey_for_workflow.py`](valkey_for_workflow.py) - Workflow with Valkey storage

cookbook/06_storage/valkey/TEST_LOG.md

@@ -0,0 +1,31 @@
+# Test Log: valkey
+
+### valkey_for_agent.py
+
+**Status:** PASS
+
+**Description:** Runs an agent backed by ValkeyDb, asks two follow-up questions with history in context, and verifies sessions are persisted via `get_sessions`.
+
+**Result:** Agent answered both turns (correctly resolved "their" from prior turn via history); `get_sessions` reported 3 agent sessions persisted in Valkey.
+
+---
+
+### valkey_for_team.py
+
+**Status:** PASS
+
+**Description:** Runs a HackerNews + web-search team backed by ValkeyDb with `output_schema=Article`, and persists team/member sessions.
+
+**Result:** Team completed and returned a valid structured `Article` (title, summary, reference_links); team session index present in Valkey. Note: the external DuckDuckGo (`ddgs`) web search intermittently failed with RemoteProtocolError/TimeoutException and was retried; total run took ~13 min. This is an upstream web-search flakiness, not a Valkey/cookbook issue.
+
+---
+
+### valkey_for_workflow.py
+
+**Status:** PASS
+
+**Description:** Runs a two-step workflow (research team then content planner) using `ValkeyDb(session_table="workflow_session")`.
+
+**Result:** Workflow completed in ~444s and produced a 4-week content plan; workflow session index (`workflow_id:content-creation-workflow`) present in Valkey.
+
+---

cookbook/06_storage/valkey/valkey_for_agent.py

@@ -0,0 +1,46 @@
+"""
+Example showing how to use Valkey as the database for an agent.
+
+Run `uv pip install valkey-glide-sync openai ddgs` to install dependencies.
+
+We can start Valkey locally using docker:
+1. Start Valkey container
+`docker run --name my-valkey -p 6379:6379 -d valkey/valkey-bundle`
+
+2. Verify container is running
+`docker ps`
+
+3. Run the file
+`python cookbook/06_storage/valkey/valkey_for_agent.py`
+"""
+
+from agno.agent import Agent
+from agno.db.base import SessionType
+from agno.db.valkey import ValkeyDb
+from agno.tools.websearch import WebSearchTools
+
+# ---------------------------------------------------------------------------
+# Setup
+# ---------------------------------------------------------------------------
+db = ValkeyDb()
+
+# ---------------------------------------------------------------------------
+# Create Agent
+# ---------------------------------------------------------------------------
+agent = Agent(
+    db=db,
+    tools=[WebSearchTools()],
+    add_history_to_context=True,
+)
+
+# ---------------------------------------------------------------------------
+# Run Agent
+# ---------------------------------------------------------------------------
+if __name__ == "__main__":
+    agent.print_response("How many people live in Canada?")
+    agent.print_response("What is their national anthem called?")
+
+    # Verify db contents
+    print("\nVerifying db contents...")
+    all_sessions = db.get_sessions(session_type=SessionType.AGENT)
+    print(f"Total sessions in Valkey: {len(all_sessions)}")

cookbook/06_storage/valkey/valkey_for_team.py

@@ -0,0 +1,77 @@
+"""
+Example showing how to use Valkey as the database for a team.
+
+Run: `uv pip install ddgs valkey-glide-sync` to install the dependencies
+
+We can start Valkey locally using docker:
+1. Start Valkey container
+`docker run --name my-valkey -p 6379:6379 -d valkey/valkey-bundle`
+
+2. Verify container is running
+`docker ps`
+
+3. Run the file
+`python cookbook/06_storage/valkey/valkey_for_team.py`
+"""
+
+from typing import List
+
+from agno.agent import Agent
+from agno.db.valkey import ValkeyDb
+from agno.models.openai import OpenAIResponses
+from agno.team import Team
+from agno.tools.hackernews import HackerNewsTools
+from agno.tools.websearch import WebSearchTools
+from pydantic import BaseModel
+
+# ---------------------------------------------------------------------------
+# Setup
+# ---------------------------------------------------------------------------
+db = ValkeyDb()
+
+
+# ---------------------------------------------------------------------------
+# Create Team
+# ---------------------------------------------------------------------------
+class Article(BaseModel):
+    title: str
+    summary: str
+    reference_links: List[str]
+
+
+hn_researcher = Agent(
+    name="HackerNews Researcher",
+    model=OpenAIResponses(id="gpt-5.5"),
+    role="Gets top stories from hackernews.",
+    tools=[HackerNewsTools()],
+)
+
+web_searcher = Agent(
+    name="Web Searcher",
+    model=OpenAIResponses(id="gpt-5.5"),
+    role="Searches the web for information on a topic",
+    tools=[WebSearchTools()],
+    add_datetime_to_context=True,
+)
+
+
+hn_team = Team(
+    name="HackerNews Team",
+    model=OpenAIResponses(id="gpt-5.5"),
+    members=[hn_researcher, web_searcher],
+    db=db,
+    instructions=[
+        "First, search hackernews for what the user is asking about.",
+        "Then, ask the web searcher to search for each story to get more information.",
+        "Finally, provide a thoughtful and engaging summary.",
+    ],
+    output_schema=Article,
+    markdown=True,
+    show_members_responses=True,
+)
+
+# ---------------------------------------------------------------------------
+# Run Team
+# ---------------------------------------------------------------------------
+if __name__ == "__main__":
+    hn_team.print_response("Write an article about the top 2 stories on hackernews")

cookbook/06_storage/valkey/valkey_for_workflow.py

@@ -0,0 +1,75 @@
+"""
+Valkey Storage for Workflow
+===========================
+
+Demonstrates using ValkeyDb as the session storage backend for a workflow.
+"""
+
+from agno.agent import Agent
+from agno.db.valkey import ValkeyDb
+from agno.models.openai import OpenAIResponses
+from agno.team import Team
+from agno.tools.hackernews import HackerNewsTools
+from agno.tools.websearch import WebSearchTools
+from agno.workflow.step import Step
+from agno.workflow.workflow import Workflow
+
+# ---------------------------------------------------------------------------
+# Create Workflow
+# ---------------------------------------------------------------------------
+hackernews_agent = Agent(
+    name="Hackernews Agent",
+    model=OpenAIResponses(id="gpt-5.5"),
+    tools=[HackerNewsTools()],
+    role="Extract key insights and content from Hackernews posts",
+)
+web_agent = Agent(
+    name="Web Agent",
+    model=OpenAIResponses(id="gpt-5.5"),
+    tools=[WebSearchTools()],
+    role="Search the web for the latest news and trends",
+)
+
+# Define research team for complex analysis
+research_team = Team(
+    name="Research Team",
+    members=[hackernews_agent, web_agent],
+    instructions="Research tech topics from Hackernews and the web",
+)
+
+content_planner = Agent(
+    name="Content Planner",
+    model=OpenAIResponses(id="gpt-5.5"),
+    instructions=[
+        "Plan a content schedule over 4 weeks for the provided topic and research content",
+        "Ensure that I have posts for 3 posts per week",
+    ],
+)
+
+# Define steps
+research_step = Step(
+    name="Research Step",
+    team=research_team,
+)
+
+content_planning_step = Step(
+    name="Content Planning Step",
+    agent=content_planner,
+)
+
+# ---------------------------------------------------------------------------
+# Run Workflow
+# ---------------------------------------------------------------------------
+if __name__ == "__main__":
+    content_creation_workflow = Workflow(
+        name="Content Creation Workflow",
+        description="Automated content creation from blog posts to social media",
+        db=ValkeyDb(
+            session_table="workflow_session",
+        ),
+        steps=[research_step, content_planning_step],
+    )
+    content_creation_workflow.print_response(
+        input="AI trends in 2024",
+        markdown=True,
+    )

cookbook/07_knowledge/05_integrations/README.md

@@ -33,6 +33,8 @@ Specific reader, cloud storage, and vector database integrations.
 | [vector_dbs/01_qdrant.py](./vector_dbs/01_qdrant.py) | Qdrant (recommended for production) |
 | [vector_dbs/02_local.py](./vector_dbs/02_local.py) | ChromaDB + LanceDB (local development) |
 | [vector_dbs/03_managed.py](./vector_dbs/03_managed.py) | Pinecone + PgVector (managed/production) |
+| [vector_dbs/04_pgvector.py](./vector_dbs/04_pgvector.py) | PgVector (PostgreSQL) |
+| [vector_dbs/05_valkey.py](./vector_dbs/05_valkey.py) | Valkey (ValkeySearch via GLIDE) |
 
 ## Running