Loading the catalogue…
Agent frameworks · LlamaIndex Inc. · US
LlamaIndex
US CLOUD Act exposureOpen sourceMIT
STAV REVIEW · LAST REVIEWED 06/21/2026
“The dominant Python RAG framework — fully Sovereign when self-hosted, but LlamaCloud's unconfirmed DPA and a CVSS 9.8 CVE history make dependency hygiene non-negotiable.”
Good for
- Teams building RAG and agent pipelines who want complete control over Inference routing — no data reaches the vendor by default
- Sovereign deployments: route against on-premise or EU-resident endpoints and the OSS library carries zero CLOUD Act exposure
- Organisations that need MIT licensing with no copyleft obligations for commercial products
- Enterprises wanting a supported path to on-premise deployment, with SOC 2 Type 2 on the commercial LlamaCloud tier
- Python-native teams who benefit from 15M+ monthly downloads worth of community integrations, connectors, and ecosystem tooling
Watch out for
- At least 11 public CVEs on record — a lower bound — including a CVSS 9.8 critical SQL injection (CVE-2025-1793, patched June 2025) exploitable via LLM-mediated prompt injection across eight vector-store integrations; pin llama-index-core versions and monitor GHSA feeds actively
- No Data Processing Agreement has been confirmed for LlamaCloud from public sources across multiple curator runs — a hard blocker for EU organisations routing personal or regulated data through the cloud tier
- Supply-chain hygiene gaps throughout: no SBOM published, PyPI binaries are unsigned, no security.txt, and no vendor-managed bug bounty — disclosures route through third-party Huntr.com
- LlamaCloud SaaS is US-hosted with full CLOUD Act exposure; any enterprise data sent to the cloud tier transits US infrastructure regardless of where your application runs
Editorial assessment, not legal advice. Stav's risk ratings, scores, and verdicts are our own analysis of publicly available information and may be incomplete or out of date. Verify independently before making compliance or procurement decisions.
MODEL REQUIREMENTS
What LlamaIndex needs from a model.
Required
None catalogued.
Recommended
- Long context (≥128k)
- Streaming
- Structured output
- Tool calling
Known issues8 open
Public security, privacy and governance advisories that are still open. Each row carries a source link — these are factual claims, not editorial judgements.
2026-06-30
lowUpstream change on llamaindex: github_release 'v0.14.22' → 'v0.14.23', github_last_push '2026-06-20T00:09:30Z' → '2026-06-26T14:23:35Z'
other · open
Source ↗2026-06-23
lowUpstream change on llamaindex: github_last_push '2026-06-15T22:21:18Z' → '2026-06-20T00:09:30Z'
other · open
Source ↗2026-06-16
lowUpstream change on llamaindex: github_last_push '2026-06-04T16:59:44Z' → '2026-06-15T22:21:18Z'
other · open
Source ↗2026-06-09
lowUpstream change on llamaindex: github_last_push '2026-05-29T21:52:59Z' → '2026-06-04T16:59:44Z'
other · open
Source ↗2026-06-02
lowUpstream change on llamaindex: github_last_push '2026-05-15T19:35:23Z' → '2026-05-29T21:52:59Z'
other · open
Source ↗