CVE-2026-50181 Langroid: Path traversal in the file tools allows read/write outside configured current directory

Langroid is a framework for building large-language-model-powered applications. Prior to version 0.64.0, Langroid's ReadFileTool and WriteFileTool appear to treat currdir as the intended working-directory boundary for file operations. However, the tools only change the process working directory t...

CVE-2026-50181

CVE-2026-50181 affects Langroid up to version 0.63.0, where ReadFileTool and WriteFileTool incorrectly trust curr_dir as a boundary. The tools change the process working directory to curr_dir and then operate on user-supplied file_path without resolving that the final path remains inside curr_dir...

CVE-2026-50180

Summary: CVE-2026-50180 in Langroid’s SQLChatAgent allows arbitrary PostgreSQL file reads due to an incomplete DANGEROUS_SQL_PATTERNS blocklist. The blocklist misses several pg_read * and related functions (e.g., pg_read_file, pg_stat_file, pg_ls_logdir, pg_ls_waldir, pg_current_logfile) and does...

CVE-2026-50180 Langroid: SQLChatAgent _validate_query blocklist misses pg_read_file family enabling arbitrary file read

Langroid is a framework for building large-language-model-powered applications. Prior to version 0.64.0, SQLChatAgent in langroid ships a validatequery defense-in-depth layer whose DANGEROUSSQLPATTERNS regex blocklist enumerates dangerous SQL primitives by specific function name. The list misses...

CVE-2026-55615

Langroid CVE-2026-55615 affects the Neo4jChatAgent path, where LLM-generated Cypher queries are sent unvalidated to the Neo4j driver prior to version 0.65.5. Connected docs describe a prompt-injection risk enabling read/write/destroy of graph data and, with APOC or dbms.security, potential OS com...

CVE-2026-55615

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...

CVE-2026-55615 Langroid: Neo4jChatAgent executes LLM-generated Cypher without validation (prompt-to-Cypher injection; config-conditional RCE), mirroring the SQLChatAgent bug fixed in CVE-2026-25879

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...

CVE-2026-55615 Langroid: Neo4jChatAgent executes LLM-generated Cypher without validation (prompt-to-Cypher injection; config-conditional RCE), mirroring the SQLChatAgent bug fixed in CVE-2026-25879

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...

Arbitrary Code Injection

Langroid is vulnerable to Arbitrary Code Injection. The vulnerability is due to improper sandboxing of Python eval that fails to remove builtins, which allows an attacker to execute arbitrary operating system commands through crafted LLM-generated tool messages...

PYSEC-2026-1532 Langroid Allows XXE Injection via XMLToolMessage

Summary A LLM application leveraging XMLToolMessage class may be exposed to untrusted XML input that could result in DoS and/or exposing local files with sensitive information. Details XMLToolMessage uses lxml without safeguards:...

CVE-2026-54771

creationtimestamp| type| source ---|---|--- 2026-07-06 22:35:22+00:00| published-proof-of-concept| https://github.com/langroid/langroid/security/advisories/GHSA-gjgq-w2m6-wr5q 2026-07-10 01:00:49+00:00| seen| https://bsky.app/profile/thehackerwire.bsky.social/post/3mqax4od5hu2i 2026-07-10...

GHSA-2PQ5-3Q89-J7CC Langroid: Neo4jChatAgent executes LLM-generated Cypher without validation (prompt-to-Cypher injection; config-conditional RCE), mirroring the SQLChatAgent bug fixed in CVE-2026-25879

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...

Improper Neutralization of Special Elements in Output Used by a Downstream Component ('Injection')

Overview langroid is a Harness LLMs with Multi-Agent Programming Affected versions of this package are vulnerable to Improper Neutralization of Special Elements in Output Used by a Downstream Component 'Injection' via the Neo4jChatAgent process. An attacker can gain unauthorized access to all gra...

GHSA-GJGQ-W2M6-WR5Q Langroid: handle_message() executes user-supplied tool JSON without sender verification

Summary A Langroid application exposing a chat interface to untrusted users may allow direct tool invocation via raw JSON payloads, even when tools are registered with use=False, handle=True. Details enablemessage..., use=False, handle=True only prevents the LLM from being instructed to generate...

Improper Neutralization of Special Elements in Output Used by a Downstream Component ('Injection')

Overview langroid is a Harness LLMs with Multi-Agent Programming Affected versions of this package are vulnerable to Improper Neutralization of Special Elements in Output Used by a Downstream Component 'Injection' via the handlemessage process. An attacker can execute arbitrary tool handlers by...

Langroid: handle_message() executes user-supplied tool JSON without sender verification

Summary A Langroid application exposing a chat interface to untrusted users may allow direct tool invocation via raw JSON payloads, even when tools are registered with use=False, handle=True. Details enablemessage..., use=False, handle=True only prevents the LLM from being instructed to generate...

Arbitrary Code Injection

Overview langroid is a Harness LLMs with Multi-Agent Programming Affected versions of this package are vulnerable to Arbitrary Code Injection in the eval process used by certain agent components when evaluating external input with fulleval=True. An attacker can execute arbitrary system commands b...

GHSA-Q9P7-WQXG-MRHC Langroid: Sandbox Escape to Remote Code Execution via Incomplete `eval()` Mitigation in TableChatAgent

Advisory Details Title: Sandbox Escape to Remote Code Execution via Incomplete eval Mitigation in TableChatAgent Description: Summary Langroid is vulnerable to a critical Sandbox Escape leading to Remote Code Execution RCE in its TableChatAgent and VectorStore capabilities. When these agents...

SQL Injection

Overview langroid is a Harness LLMs with Multi-Agent Programming Affected versions of this package are vulnerable to SQL Injection through the SQLChatAgent process. An attacker can access arbitrary files on the server by crafting SQL queries that bypass the regex-based blocklist using quoted...

PT-2026-56064

Name of the Vulnerable Software and Affected Versions Langroid versions prior to 0.65.3 Description An application using this framework that exposes a chat interface to untrusted users may allow direct tool invocation via raw JSON payloads. This occurs even when tools are registered with use=Fals...