CVEs / CVE-2026-114510
CVE-2026-114510
Flexible GraphRAG is a Python framework for building graph and retrieval-augmented generation applications across multiple graph, vector, and search backends. From 0.6.0 until 0.8.2, the _escape() helper in flexible-graphrag/langchain/graph/pg_store_adapters/falkordb_adapter.py escapes quotes before backslashes and then interpolates the result into Cypher assembled by FalkorDBAdapter.add_graph_documents, allowing an inserted escape to be doubled and a quote to terminate the string early. When GRAPH_BACKEND=langchain and GRAPH_DB=falkordb are configured, attacker-influenced ingested document content can cause the extraction model to produce malicious entity or relationship values that reach these queries, allowing injected Cypher clauses to modify or delete graph data and, when the calling code surfaces query results to the user or LLM, disclose graph data. Dynamic node and relationship type identifiers are also inserted into backtick-quoted query positions without safe parameterization. The LlamaIndex FalkorDB store is not affected. This issue is fixed in version 0.8.2.
Affected products · 2
| Product | Versions affected | Fixed in |
|---|---|---|
| LangChain | not stated by CPE | — |
| LlamaIndex | not stated by CPE | — |
Versions as NVD's CPE configuration states them; "—" means the record gives no fixed version, not that none exists. Check the vendor's advisory in the references.
In this registry
- LangChain — agent dossier
- LangChain Deep Agents — agent dossier
- LlamaIndex — agent dossier
References · 3
- github.com/stevereiner/flexible-graphrag/commit/9ace8d47e198ce3f15dbf22853e396ab4e9884fa
- github.com/stevereiner/flexible-graphrag/releases/tag/v0.8.2
- github.com/stevereiner/flexible-graphrag/security/advisories/GHSA-vqrg-773h-4wxc
The official record's references, as NVD lists them, minus any tagged Exploit. Nothing here is an exploit or a proof of concept.
Description, scores, versions and references are the official CVE record's own, retrieved 2026-10-25 from NVD. Category, attack surface and cross-links are this registry's classification and can be wrong — use the button above. Not a severity assessment for your deployment.