LlamaIndex
The leading data framework for LLM apps — 150+ data loaders, RAG pipelines, and agent tools for connecting any data source to any model.
LlamaIndex
LlamaIndex (formerly GPT Index) is the most widely-used data framework for building LLM-powered applications over your own data. It provides over 150 data loaders, composable RAG pipelines, and agent tool integrations so you can connect any data source — PDFs, databases, APIs, Notion, Slack, and more — to any LLM.
Key features
- 150+ data loaders — ingest PDFs, DOCX, HTML, CSV, Notion, Google Drive, Slack, GitHub, and more
- Composable RAG pipelines — chunking, embedding, indexing, retrieval, and synthesis as modular components
- Agent tools — wrap indices as tools for ReAct, OpenAI function-calling, or custom agents
- Multiple index types — vector, keyword, list, tree, and knowledge graph indices
- Streaming and async — first-class async support for production workloads
- 300+ integrations — LLMs, embedding models, vector stores, and observability tools
Quick start
npx ai-supply add llama-index-data-framework
# Or install directly
pip install llama-index
from llama_index.core import VectorStoreIndex, SimpleDirectoryReader
# Load docs and build an index
docs = SimpleDirectoryReader("./data").load_data()
index = VectorStoreIndex.from_documents(docs)
# Query it
query_engine = index.as_query_engine()
response = query_engine.query("What is the main theme of these documents?")
print(response)
Curated mirror of the open-source LlamaIndex project (MIT). Install upstream from the repository.
Compromise signals — malicious or tampered code (leaked secrets, backdoors, a dropped executable) — reduce the score, and known dependency CVEs carry a bounded penalty (they warrant review but never QUARANTINE — update the dependency to clear). Other dangerous-by-capability traits are risk surface, expected for some capabilities. Every finding is mapped to its OWASP control below.
Findings mapped to the OWASP Top 10 for LLM Applications (2025) and the OWASP Machine Learning Security Top 10. Expand any flagged control for the exact findings — compromise reduces the score; expected/risk-surface do not, except a known CVE, which carries a small bounded penalty (high/critical → Review).
The same gate an agent runs before installing (POST /api/v1/trust/llama-index-data-framework/check). Click a policy:
Consume LlamaIndex programmatically. Authenticate with an API key or session — see Authorize an agent.
# Agents: CHECK BEFORE YOU INSTALL (no auth) — score, grade, level, capability manifest
curl https://ai-supply.store/api/v1/trust/llama-index-data-framework
# Gate against your org policy (returns { pass, violations })
curl -X POST https://ai-supply.store/api/v1/trust/llama-index-data-framework/check \
-H "Content-Type: application/json" \
-d '{"minGrade":"B","denyPermissions":["shell"],"denyUnknownEgress":true}'
# CLI
npx ai-supply add llama-index-data-framework
# REST (install → download)
curl -X POST https://ai-supply.store/api/v1/listings/llama-index-data-framework/install \
-H "Authorization: Bearer $AIM_KEY"
# MCP tool
install_listing({ "slug": "llama-index-data-framework" })OpenAPI spec →Curated mirror — latest upstream source. See the repository for tagged releases.