Knowledge Base

Advanced knowledge management system for AI agents

Knowledge Base System

The knowledge system provides sophisticated document processing and semantic search capabilities, enabling agents to access and utilise information from various sources.

Key Features

Process PDFs, documents, spreadsheets, images, and raw text

Multiple strategies for optimal text segmentation

Vector-based search with optional reranking

User, agent, and run-specific knowledge scoping

Optional relationship extraction and storage

Automatic quality assessment for stored knowledge

Quick Start

Agent with Knowledge

from praisonaiagents import Agent

agent = Agent(
    name="Research Assistant",
    instructions="Answer questions using the knowledge base.",
    knowledge={
        "sources": ["research_paper.pdf", "data.txt"],
        "vector_store": {
            "provider": "chroma",
            "config": {
                "collection_name": "research_docs",
                "path": ".praison"
            }
        }
    }
)

response = agent.start("What are the key findings?")

Configuration Options

Basic Configuration

knowledge_config = {
    "vector_store": {
        "provider": "chroma",
        "config": {
            "collection_name": "knowledge_base",
            "path": ".praison",
            "distance_metric": "cosine"
        }
    },
    "embedder": {
        "provider": "openai",
        "config": {
            "model": "text-embedding-3-small"
        }
    }
}

Advanced Configuration with Graph Store

knowledge_config = {
    "vector_store": {
        "provider": "chroma",
        "config": {
            "collection_name": "knowledge_base",
            "path": ".praison"
        }
    },
    "graph_store": {
        "provider": "neo4j",
        "config": {
            "url": "bolt://localhost:7687",
            "username": "neo4j",
            "password": "password"
        }
    },
    "llm": {
        "provider": "openai",
        "config": {
            "model": "gpt-4o-mini",
            "temperature": 0
        }
    },
    "reranker": {
        "enabled": True,
        "default_rerank": False
    }
}

Chunking Strategies

The chosen strategy applies to every file type, including plain-text files.

Token-based Chunking

kb = Knowledge({
    "chunker": {
        "name": "token",
        "chunk_size": 500,
        "chunk_overlap": 50
    }
})

Document Processing

Supported File Types

• PDF (.pdf)
• Word (.doc, .docx)
• Text (.txt)
• Markdown (.md)
• RTF (.rtf)

• Excel (.xls, .xlsx)
• CSV (.csv)
• JSON (.json)
• XML (.xml)

• Images (OCR)
• Local HTML files (.html, .htm)
• Raw text strings
• YouTube videos
• Remote URLs (http:// / https://) — via MarkItDown, Knowledge.add(url)

All supported UTF-8 text extensions — .txt, .md, .csv, .json, .xml, .html, .htm — are chunked with the configured chunker (respecting chunk_size and chunk_overlap), and the original case is preserved. Behaviour matches every other file type. Since PraisonAI PR #4831.

Processing Options

# Add with metadata
kb.add(
    "research.pdf",
    user_id="user123",
    metadata={
        "category": "AI Research",
        "year": 2024,
        "author": "Dr. Smith"
    }
)

# Batch processing
documents = ["doc1.pdf", "doc2.txt", "doc3.md"]
for doc in documents:
    kb.add(doc, user_id="user123")

As of PraisonAI #4334 (2026-08-25), Knowledge.add() accepts remote URLs directly. The URL is fetched and extracted with MarkItDown (already a Praison runtime dependency), chunked with the configured chunker, and stored on the vector store — the same pipeline used for local files. Stored chunks carry metadata["source_type"] = "url" and metadata["source"] = <the URL>. A single bad URL in a batch surfaces the error rather than being silently dropped, matching local-file behaviour.

from praisonaiagents import Knowledge

kb = Knowledge(config={
    "vector_store": {"provider": "chroma", "config": {"collection_name": "web_docs"}}
})

kb.add("https://example.com/article", user_id="user123")
kb.add("https://arxiv.org/pdf/2301.00000.pdf", user_id="user123")   # PDFs via URL work too

# Metadata is merged with the auto-added source_type/source keys:
kb.add(
    "https://blog.example.com/post",
    user_id="user123",
    metadata={"category": "blog", "year": 2026},
)

results = kb.search("what does the article say?", user_id="user123")

Agent(knowledge=["https://..."]) is a separate code path and still skips remote URLs (PR #4004). Build a shared Knowledge instance, call kb.add(url), then pass knowledge=kb to the agent. See Knowledge Quick Start.

Search Features

Basic Search

# Simple search
results = kb.search("artificial intelligence", limit=5)

# User-scoped search
results = kb.search(
    query="machine learning",
    user_id="user123",
    limit=10
)

Advanced Search Options

With Reranking

# Enable Mem0 reranking for better relevance
results = kb.search(
    query="neural networks",
    user_id="user123",
    rerank=True,
    top_k=20  # Retrieve more before reranking
)

Memory Integration

When used with agents, knowledge automatically integrates with memory:

agent = Agent(
    name="Research Assistant",
    knowledge={
        "sources": ["papers/"],
        "vector_store": {
            "provider": "chroma",
            "config": {"collection_name": "research_docs"}
        }
    },
    memory=True  # Enable memory integration
)

# Knowledge is automatically searched during conversations
response = agent.chat("What does the research say about transformers?")

Graph Store Features

Graph stores enable relationship extraction and complex queries beyond simple semantic search.

Configuration

knowledge_config = {
    "graph_store": {
        "provider": "neo4j",  # or "memgraph"
        "config": {
            "url": "bolt://localhost:7687",
            "username": "neo4j",
            "password": "password"
        }
    },
    "extract_relationships": True
}

Relationship Queries

# Find related concepts
results = kb.search_graph(
    "What concepts are related to transformers?",
    user_id="user123"
)

# Explore connections
results = kb.search_graph(
    "How is attention mechanism connected to BERT?",
    user_id="user123"
)

Best Practices

Chunking Strategy

• Smaller chunks (100-200 tokens): Better precision
• Larger chunks (500-1000 tokens): Better context
• Match chunk size to query complexity

Organisation

• Separate collections by domain
• Use metadata for filtering
• Regular cleanup of outdated content

Performance

• Enable caching for repeated queries
• Use appropriate embedding models
• Batch document processing

Quality

• Verify document processing
• Monitor search relevance
• Regular reindexing if needed

Example: Research Assistant

Complete Example

from praisonaiagents import Agent
from praisonaiagents import Knowledge

# Configure knowledge base
knowledge_config = {
    "vector_store": {
        "provider": "chroma",
        "config": {
            "collection_name": "research_papers",
            "path": ".praison"
        }
    },
    "chunker": {
        "name": "semantic",
        "threshold": 0.7
    },
    "embedder": {
        "provider": "openai",
        "config": {
            "model": "text-embedding-3-small"
        }
    },
    "reranker": {
        "enabled": True
    }
}

# Create research assistant
research_agent = Agent(
    name="Research Assistant",
    instructions="""You are an expert research assistant.
    Use the knowledge base to provide accurate, well-sourced answers.
    Always cite the specific documents you reference.""",
    knowledge={
        "sources": ["research_papers"],
        **knowledge_config
    }
)

# Use the assistant
response = research_agent.chat(
    "What are the main approaches to AI alignment?"
)

# Direct knowledge queries
kb = research_agent.knowledge_instance
papers = kb.search("alignment techniques", limit=5)
for paper in papers:
    print(f"- {paper['text'][:100]}...")
    print(f"  Source: {paper.get('metadata', {}).get('source')}")

Troubleshooting

Symptom	Likely cause	Fix
RuntimeError: Chroma persist failed at '<path>' (PanicException: ...) on Agent.start(...) or Knowledge.add(...)	The on-disk sqlite Chroma store at <path> is corrupt, locked by another process, or was written by an incompatible Chroma version.	Delete <path> and re-index, or point knowledge={"vector_store": {"config": {"path": "/a/fresh/dir"}}} at a fresh directory.
RuntimeError: Knowledge indexing backend crashed on <file> while indexing a directory	The vector-store backend panicked on that file (typically a Chroma rust-binding panic on a corrupt store).	Delete the persist directory or move <file> aside and re-run.
Two unrelated projects appear to share the same Knowledge base	You are on a pre-#4422 release where the default path was the cwd-relative string "knowledge_db".	Upgrade to the release built from PraisonAI PR #4422; defaults are now absolute (<project>/.praisonai/knowledge/chroma).
Windows: Agent(knowledge=["docs/"]) SIGSEGVs or hangs	Pre-#4422 issue #4376: pyo3 PanicException (a BaseException) unwound the process instead of surfacing as a Python error.	Upgrade; panics are now caught and re-raised as RuntimeError with an actionable message.

Next Steps

Learn about memory integration

Build RAG applications