https://cve.radiocsirt.org/rss/recent/all/10 Most recent entries from all 2026-10-03T09:32:10.425746+00:00 Vulnerability-Lookup csirt@opendfir.org python-feedgen Contains only the most 10 recent entries. https://cve.radiocsirt.org/vuln/cve-2026-55615 CVE-2026-55615 — Langroid: Neo4jChatAgent executes LLM-generated Cypher without validation (prompt-to-Cypher injection; config-condition… 2026-10-03T09:32:10.457680+00:00

Affected: langroid

Langroid is a framework for building large-language-model-powered applications. Prior to version 0.65.5, Neo4jChatAgent passes LLM-generated Cypher queries straight to the Neo4j driver with no validation, no statement-type allowlist, and no opt-out gate. The query text is influenceable by prompt injection (direct user input or indirect content the agent reads back via RAG), so an attacker who can influence the prompt can read or destroy all graph data and, when APOC or dbms.security procedures are enabled on the server, achieve OS-command and filesystem access. This is the same defect class and threat model as the SQLChatAgent prompt-to-SQL-to-RCE issue fixed in version 0.63.0 (CVE-2026-25879); that fix did not extend to the neo4j module. Version 0.65.5 contains a fix for the neo4j module.

 https://cve.radiocsirt.org/vuln/ghsa-2pq5-3q89-j7cc GHSA-2pq5-3q89-j7cc — Langroid: Neo4jChatAgent executes LLM-generated Cypher without validation (prompt-to-Cypher injection; config-condition… 2026-10-03T09:32:10.457745+00:00

Affected: PyPI: langroid

Neo4jChatAgent passes LLM-generated Cypher queries straight to the Neo4j driver with no validation, no statement-type allowlist, and no opt-out gate. The query text is influenceable by prompt injection (direct user input or indirect content the agent reads back via RAG), so an attacker who can influence the prompt can read or destroy all graph data and, when APOC or dbms.security procedures are enabled on the server, achieve OS-command and filesystem access. This is the same defect class and threat model as the SQLChatAgent prompt-to-SQL-to-RCE issue fixed in version 0.63.0 (CVE-2026-25879); that fix did not extend to the neo4j module.

## Technical detail Untrusted-input to sink trace (reviewed on langroid HEAD b9df06f, v0.65.3):

1. Tool schemas accept raw query text from the LLM. `langroid/agent/special/neo4j/tools.py:4-9` (CypherRetrievalTool.cypher_query: str) and `:15-21` (CypherCreationTool.cypher_query: str). These tools are enabled unconditionally in `neo4j_chat_agent.py:412-419` (enable_message([GraphSchemaTool, CypherRetrievalTool, CypherCreationTool, DoneTool])).

2. Read path. `neo4j_chat_agent.py:300` cypher_retrieval_tool(msg) -> `:325` query = msg.cypher_query -> `:328` self.read_query(query) -> `:223` session.run(query, parameters). The LLM-controlled string is the first positional argument to session.run; parameters is None. No validation occurs between :325 and :223.

3. Write path. `neo4j_chat_agent.py:338` cypher_creation_tool(msg) -> `:348` query = msg.…