CVE-2026-54771
Langroid is a framework for building large-language-model-powered applications. Prior to version 0.65.3, 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`. Version 0.65.3 fixes the issue.
Scoring
- Severity
- HIGH
- CVSS base score
- 8.1
- CVSS vector
- CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:N
- EPSS probability
- 0.39%
- CWE
- CWE-75
- Published
- 2026-07-09
- Last modified
- 2026-07-10
Affected products
- langroid langroid
Weakness type
Related vulnerabilities
- CVE-2024-58362 — SurrealDB before 1.5.5 Query Injection via RPC API
- CVE-2026-31908 — Apache APISIX: forward auth plugin allows header injection
- CVE-2026-29042 — Nuclio Shell Runtime Command Injection Leading to Privilege Escalation
- CVE-2026-27120 — Leaf-kit html escaping does not work on characters that are part of extended grapheme cluster
- CVE-2025-61911 — python-ldap has sanitization bypass in ldap.filter.escape_filter_chars
- CVE-2025-50213 — Apache Airflow Providers Snowflake: Potential SQL injection in CopyFromExternalStageToSnowflakeOperator
- CVE-2024-9940 — Calculated Fields Form <= 5.2.45 - HTML Injection
- CVE-2023-1758 — Failure to Sanitize Special Elements into a Different Plane (Special Element Injection) in thorsten/phpmyfaq