"""
ChromaDB with Hybrid Search using Reciprocal Rank Fusion (RRF)
This example demonstrates how to use ChromaDB with hybrid search,
which combines dense vector similarity search (semantic) with
full-text search (keyword/lexical) using RRF fusion.
Hybrid search is useful when you want to:
- Combine semantic understanding with exact keyword matching
- Improve retrieval accuracy for queries with specific terms
- Handle both conceptual and lexical search needs
The RRF algorithm fuses rankings from both search methods using:
RRF(d) = sum(1 / (k + rank_i(d))) for each ranking i
"""
import asyncio
from textwrap import dedent
from agno.agent import Agent
from agno.knowledge.embedder.openai import OpenAIEmbedder
from agno.knowledge.knowledge import Knowledge
from agno.models.openai import OpenAIChat
from agno.vectordb.chroma import ChromaDb, SearchType
# ---------------------------------------------------------------------------
# Create Knowledge Base
# ---------------------------------------------------------------------------
knowledge = Knowledge(
name="Agno Documentation",
description="Knowledge base for Agno framework documentation",
vector_db=ChromaDb(
name="agno_docs",
path="tmp/chromadb_hybrid",
persistent_client=True,
# Enable hybrid search - combines vector similarity with keyword matching using RRF
search_type=SearchType.hybrid,
# RRF (Reciprocal Rank Fusion) constant - controls ranking smoothness.
# Higher values (e.g., 60) give more weight to lower-ranked results,
# Lower values make top results more dominant. Default is 60 (per original RRF paper).
hybrid_rrf_k=60,
embedder=OpenAIEmbedder(id="text-embedding-3-small"),
),
max_results=10,
)
# ---------------------------------------------------------------------------
# Create Agent
# ---------------------------------------------------------------------------
description = dedent(
"""\
You are AgnoAssist — an AI Agent built to help developers learn and master the Agno framework.
Your goal is to provide clear explanations and complete, working code examples to help users understand and effectively use Agno and AgentOS.\
"""
)
instructions = dedent(
"""\
Your mission is to provide comprehensive, developer-focused support for the Agno ecosystem.
Follow this structured process to ensure accurate and actionable responses:
1. **Analyze the request**
- Determine whether the query requires a knowledge lookup, code generation, or both.
- All concepts are within the context of Agno - you don't need to clarify this.
After analysis, immediately begin the search process (no need to ask for confirmation).
2. **Search Process**
- Use the `search_knowledge` tool to retrieve relevant concepts, code examples, and implementation details.
- Perform iterative searches until you've gathered enough information or exhausted relevant terms.
Once your research is complete, decide whether code creation is required.
If it is, ask the user if they'd like you to generate an Agent for them.
3. **Code Creation**
- Provide fully working code examples that can be run as-is.
- Always use `agent.run()` (not `agent.print_response()`).
- Include all imports, setup, and dependencies.
- Add clear comments, type hints, and docstrings.
- Demonstrate usage with example queries.
Example:
~~~python
from agno.agent import Agent
from agno.tools.websearch import WebSearchTools
agent = Agent(tools=[WebSearchTools()])
response = agent.run("What's happening in France?")
print(response)
