LlamaIndex Agentic Systems

Build production-grade agentic RAG systems with semantic ingestion, knowledge graphs, dynamic routing, and observability.

Quick Start

Build a working agent in 6 steps:

Step 1: Install Dependencies

bash

pip install llama-index-core>=0.10.0 llama-index-llms-openai llama-index-embeddings-openai arize-phoenix

See scripts/requirements.txt for full pinned dependencies.

Step 2: Ingest with Semantic Chunking

python

from llama_index.core import SimpleDirectoryReader
from llama_index.core.node_parser import SemanticSplitterNodeParser
from llama_index.embeddings.openai import OpenAIEmbedding

embed_model = OpenAIEmbedding(model_name="text-embedding-3-small")
splitter = SemanticSplitterNodeParser(
    buffer_size=1,
    breakpoint_percentile_threshold=95,
    embed_model=embed_model
)

docs = SimpleDirectoryReader(input_files=["data.pdf"]).load_data()
nodes = splitter.get_nodes_from_documents(docs)

Step 3: Build Index

python

from llama_index.core import VectorStoreIndex

index = VectorStoreIndex(nodes, embed_model=embed_model)
index.storage_context.persist(persist_dir="./storage")

Step 4: Verify Index

python

# Confirm index built correctly
print(f"Indexed {len(index.docstore.docs)} document chunks")

# Preview a sample node
sample = list(index.docstore.docs.values())[0]
print(f"Sample chunk: {sample.text[:200]}...")

Step 5: Create Query Engine

python

query_engine = index.as_query_engine(similarity_top_k=5)
response = query_engine.query("What are the key concepts?")
print(response)

Step 6: Enable Observability

python

import phoenix as px
import llama_index.core

px.launch_app()
llama_index.core.set_global_handler("arize_phoenix")
# All subsequent queries are now traced

For production script, run: python scripts/ingest_semantic.py

Architecture Overview

Six pillars for agentic systems:

Pillar	Purpose	Reference
Ingestion	Semantic chunking, code splitting, metadata	references/ingestion.md
Retrieval	BM25 keyword search, hybrid fusion	references/retrieval-strategies.md
Property Graphs	Knowledge graphs + vector hybrid	references/property-graphs.md
Context RAG	Query routing, decomposition, reranking	references/context-rag.md
Orchestration	ReAct agents, event-driven Workflows	references/orchestration.md
Observability	Tracing, debugging, evaluation	references/observability.md

Decision Trees

Which Node Parser?

code

Is the content source code?
├─ Yes → CodeSplitter
│        language="python" (or typescript, javascript, java, go)
│        chunk_lines=40, chunk_lines_overlap=15
│        → See: references/ingestion.md#codesplitter
│
└─ No, it's documents:
    ├─ Need semantic coherence (legal, technical docs)?
    │   └─ Yes → SemanticSplitterNodeParser
    │            buffer_size=1 (sensitive), 3 (stable)
    │            breakpoint_percentile_threshold=95 (fewer), 70 (more)
    │            → See: references/ingestion.md#semanticsplitternodeparser
    │
    ├─ Prioritize speed → SentenceSplitter
    │        chunk_size=1024, chunk_overlap=20
    │        → See: references/ingestion.md#sentencesplitter
    │
    └─ Need fine-grained retrieval → SentenceWindowNodeParser
             window_size=3 (surrounding sentences in metadata)
             → See: references/ingestion.md#sentencewindownodeparser

Trade-off: Semantic chunking requires embedding calls during ingestion (cost + latency).

Which Retrieval Mode?

code

Query contains exact terms (function names, error codes, IDs)?
├─ Yes, exact match critical → BM25
│        retriever = BM25Retriever.from_defaults(nodes=nodes)
│        → See: references/retrieval-strategies.md#bm25retriever
│
├─ Conceptual/semantic query → Vector
│        retriever = index.as_retriever(similarity_top_k=5)
│        → See: references/context-rag.md
│
└─ Mixed or unknown query type → Hybrid (recommended default)
         alpha=0.5 (equal weight), 0.3 (favor BM25), 0.7 (favor vector)
         → See: references/retrieval-strategies.md#hybrid-search

Trade-off: Hybrid adds BM25 index overhead but provides most robust retrieval.

Which Graph Extractor?

code

Need document navigation only (prev/next/parent)?
├─ Yes → ImplicitPathExtractor (no LLM, zero cost)
│        → See: references/property-graphs.md#implicitpathextractor
│
└─ No, need semantic relationships:
    ├─ Fixed ontology required (regulated domain)?
    │   └─ Yes → SchemaLLMPathExtractor
    │            Pass schema: {"PERSON": ["WORKS_AT"], "COMPANY": ["LOCATED_IN"]}
    │            → See: references/property-graphs.md#schemallmpathextractor
    │
    └─ No, discovery/exploration:
        └─ SimpleLLMPathExtractor
           max_paths_per_chunk=10 (control noise)
           → See: references/property-graphs.md#simplellmpathextractor

Which Graph Retriever?

code

Need SQL-like aggregations (COUNT, SUM)?
├─ Yes, trusted environment → TextToCypherRetriever
│        Risk: LLM syntax errors, injection
│        → See: references/property-graphs.md#texttocypherretriever
│
├─ Yes, need safety → CypherTemplateRetriever
│        Pre-define: MATCH (p:Person {name: $name}) RETURN p
│        LLM only extracts parameters
│        → See: references/property-graphs.md#cyphertemplateretriever
│
└─ No, robustness priority → VectorContextRetriever
         Vector search → graph traversal (path_depth=2)
         Most reliable, no code generation
         → See: references/property-graphs.md#vectorcontextretriever

Which Agent Pattern?

code

Simple tool loop sufficient?
├─ Yes → ReAct Agent (FunctionCallingAgent)
│        Tools via FunctionTool or ToolSpec
│        → See: references/orchestration.md#react-agent-pattern
│
└─ No, need:
    ├─ Branching/cycles → Workflow
    │   → See: references/orchestration.md#branching
    ├─ Human-in-the-loop → Workflow (suspend/resume)
    │   → See: references/orchestration.md#human-in-the-loop
    ├─ Multi-agent handoff → Workflow + Concierge pattern
    │   → See: references/orchestration.md#concierge-multi-agent
    └─ Parallel execution → Workflow with multiple event emissions
        → See: references/orchestration.md#workflows

Common Patterns

Pattern 1: Metadata-Enriched Ingestion

python

from llama_index.core.extractors import TitleExtractor, SummaryExtractor, KeywordExtractor
from llama_index.core.ingestion import IngestionPipeline

pipeline = IngestionPipeline(
    transformations=[
        splitter,
        TitleExtractor(),
        SummaryExtractor(),
        KeywordExtractor(keywords=5),
        embed_model,
    ]
)
nodes = pipeline.run(documents=docs)

Pattern 2: PropertyGraphIndex with Hybrid Retrieval

python

from llama_index.core import PropertyGraphIndex
from llama_index.core.indices.property_graph import SimpleLLMPathExtractor

index = PropertyGraphIndex.from_documents(
    docs,
    embed_model=embed_model,
    kg_extractors=[SimpleLLMPathExtractor(max_paths_per_chunk=10)],
)

# Hybrid: vector search + graph traversal
retriever = index.as_retriever(include_text=True)

Pattern 3: Router with Multiple Engines

python

from llama_index.core.query_engine import RouterQueryEngine
from llama_index.core.selectors import LLMSingleSelector
from llama_index.core.tools import QueryEngineTool

tools = [
    QueryEngineTool.from_defaults(
        query_engine=summary_engine,
        description="High-level summaries and overviews"
    ),
    QueryEngineTool.from_defaults(
        query_engine=detail_engine,
        description="Specific facts, numbers, and details"
    ),
]

router = RouterQueryEngine(
    selector=LLMSingleSelector.from_defaults(),
    query_engine_tools=tools,
)

Pattern 4: Event-Driven Workflow

python

from llama_index.core.workflow import Workflow, step, StartEvent, StopEvent, Event

class QueryEvent(Event):
    query: str

class MyAgent(Workflow):
    @step
    async def classify(self, ev: StartEvent) -> QueryEvent:
        return QueryEvent(query=ev.get("query"))

    @step
    async def respond(self, ev: QueryEvent) -> StopEvent:
        result = self.query_engine.query(ev.query)
        return StopEvent(result=str(result))

# Run
agent = MyAgent(timeout=60)
result = await agent.run(query="What is X?")

Pattern 5: Reranking Pipeline

python

from llama_index.core.postprocessor import SimilarityPostprocessor, LLMRerank

query_engine = index.as_query_engine(
    similarity_top_k=10,  # Retrieve more
    node_postprocessors=[
        SimilarityPostprocessor(similarity_cutoff=0.7),
        LLMRerank(top_n=3),  # Rerank to top 3
    ]
)

Script Reference

Script	Purpose	Usage
`scripts/ingest_semantic.py`	Build index with semantic chunking + graph	`python scripts/ingest_semantic.py --doc path/to/file.pdf`
`scripts/agent_workflow.py`	Event-driven agent template	`python scripts/agent_workflow.py`
`scripts/requirements.txt`	Pinned dependencies	`pip install -r scripts/requirements.txt`

Adapt scripts by modifying configuration variables at the top of each file.

Reference Index

Load references based on task:

Task	Load Reference
Configure chunking strategy	references/ingestion.md
Add metadata extractors	references/ingestion.md
Build knowledge graph	references/property-graphs.md
Choose graph store (Neo4j, etc.)	references/property-graphs.md
Implement query routing	references/context-rag.md
Decompose complex queries	references/context-rag.md
Add reranking	references/context-rag.md
Build ReAct agent	references/orchestration.md
Create Workflow	references/orchestration.md
Multi-agent system	references/orchestration.md
Setup Phoenix tracing	references/observability.md
Debug retrieval failures	references/observability.md
Evaluate agent quality	references/observability.md

Troubleshooting

Agent says "I don't know" with relevant data

Diagnose:

bash

# Open Phoenix UI at http://localhost:6006
# Navigate to Traces → Select query → Retrieval span → Retrieved Nodes

Fix:

python

# 1. Increase retrieval candidates
query_engine = index.as_query_engine(similarity_top_k=10)  # was 5

# 2. Add reranking to improve precision
from llama_index.core.postprocessor import LLMRerank
query_engine = index.as_query_engine(
    similarity_top_k=10,
    node_postprocessors=[LLMRerank(top_n=3)]
)

Verify: Re-run query, check Phoenix shows improved relevance scores (>0.7).

Semantic chunking too slow

Diagnose:

python

# Time the ingestion
import time
start = time.time()
nodes = splitter.get_nodes_from_documents(docs)
print(f"Chunking took {time.time() - start:.1f}s for {len(docs)} docs")

Fix:

python

# Option 1: Use local embeddings (no API calls)
from llama_index.embeddings.huggingface import HuggingFaceEmbedding
embed_model = HuggingFaceEmbedding(model_name="BAAI/bge-small-en-v1.5")

# Option 2: Hybrid strategy for large corpora
bulk_nodes = SentenceSplitter().get_nodes_from_documents(bulk_docs)
critical_nodes = SemanticSplitterNodeParser(...).get_nodes_from_documents(critical_docs)

Verify: Re-run with show_progress=True, confirm <1s per document.

Graph extraction producing noise

Diagnose:

python

# Check extracted triples
for node in index.property_graph_store.get_triplets():
    print(node)  # Look for irrelevant or duplicate relationships

Fix:

python

# Option 1: Reduce paths per chunk
SimpleLLMPathExtractor(max_paths_per_chunk=5)  # was 10

# Option 2: Use strict schema
SchemaLLMPathExtractor(
    possible_entities=["PERSON", "COMPANY"],
    possible_relations=["WORKS_AT", "FOUNDED"],
    strict=True
)

Verify: Re-index, confirm triplet count reduced and relationships are relevant.

Workflow step not triggering

Diagnose:

python

# Enable verbose mode
agent = MyWorkflow(timeout=60, verbose=True)
result = await agent.run(query="test")
# Check console for: [Step Name] Received event: EventType

Fix:

python

# Verify type hints match exactly
class MyEvent(Event):
    query: str

@step
async def my_step(self, ev: MyEvent) -> StopEvent:  # Type hint must be MyEvent
    ...

Verify: Verbose output shows [my_step] Received event: MyEvent.

Phoenix not showing traces

Diagnose:

python

import phoenix as px
session = px.launch_app()
print(f"Phoenix URL: {session.url}")  # Should print http://localhost:6006

Fix:

python

# MUST call BEFORE any LlamaIndex imports/operations
import phoenix as px
px.launch_app()

import llama_index.core
llama_index.core.set_global_handler("arize_phoenix")

# Now import and use LlamaIndex
from llama_index.core import VectorStoreIndex

Verify: Make a query, refresh Phoenix UI, trace appears within 5 seconds.

When Not to Use This Skill

This skill is specific to LlamaIndex in Python. Do not use for:

•LangChain projects — Different framework, different APIs
•Pure vector search without agents — Simpler solutions exist
•Non-Python environments — All examples are Python 3.9+
•Local-only / offline setups — Scripts default to OpenAI APIs; modification required for local models
•Simple Q&A bots — Overkill if you don't need graphs, routing, or workflows

If unsure: Check if your use case involves semantic chunking, knowledge graphs, query routing, or multi-step agents. If yes, this skill applies.

Glossary

Term	Definition
Node	Chunk of text with metadata, the atomic unit of retrieval
PropertyGraphIndex	Index combining vector embeddings with labeled property graph
Extractor	Component that generates graph triples from text
Retriever	Component that fetches relevant nodes/context
Postprocessor	Filters or reranks nodes after retrieval
Workflow	Event-driven state machine for agent orchestration
Span	Duration-tracked operation in observability