Bug 2565749 (CVE-2026-116691)
| Summary: | CVE-2026-116691 mlflow: MLflow: Information disclosure via unsanitized CrewAI autologging telemetry | ||
|---|---|---|---|
| Product: | [Other] Security Response | Reporter: | OSIDB Bzimport <bzimport> |
| Component: | vulnerability | Assignee: | Product Security <prodsec-ir-bot> |
| Status: | NEW --- | QA Contact: | |
| Severity: | medium | Docs Contact: | |
| Priority: | medium | ||
| Version: | unspecified | CC: | ilpinto, ltomasbo, ykashtan |
| Target Milestone: | --- | Keywords: | Security |
| Target Release: | --- | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | --- | |
| Doc Text: |
A flaw was found in MLflow. This vulnerability can lead to sensitive information disclosure when using the CrewAI autologging integration, which serializes agent objects into trace telemetry without sanitizing credential fields. Consequently, plaintext application programming interface (API) keys for Large Language Model (LLM) providers are stored in trace spans. An attacker or unauthorized user with read access to the trace backend or export destination can obtain these credentials and gain unauthorized access to external provider accounts.
|
Story Points: | --- |
| Clone Of: | Environment: | ||
| Last Closed: | Type: | --- | |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
|
Description
OSIDB Bzimport
2026-11-03 12:26:46 UTC
|