search
MLflow
Trends
- 1Anthropic Adds Native Claude Tracing to Databricks MLflowโผAnthropic has added native Claude trace support to Databricks MLflow. Claude calls now emit structured spans that captur
Anthropic has added native Claude trace support to Databricks MLflow. Claude calls now emit structured spans capturing prompts, responses, tool calls, and timings, letting developers replay failing agent steps instead of guessing at what went wrong. The integration bundles debugging, evaluation, and monitoring into MLflow's existing workflow, and is being welcomed by engineers building Claude-powered agents.
- 2MLflow vulnerability allows remote code execution via pickle deserializationโผ๐จ EUVD-2026-116619 ๐ Score: 7.8/10 (CVSS v3.1) ๐ฆ Product: MLflow ๐ข Vendor: mlflow ๐ Updated: 2026-11-04 ๐ MLflow PickleE
A new vulnerability, EUVD-2026-116619, has been documented in MLflow, an open-source machine learning platform. The flaw, rated 7.8 out of 10 on the CVSS v3.1 scale, involves insecure deserialization of PickleEvaluationArtifact data, which could let remote attackers execute arbitrary code on affected systems. The advisory was updated on November 4, 2026, and security teams monitoring MLflow deployments are being urged to review their exposure and apply mitigations.
- 3New MLflow vulnerability allows information disclosure, scored 5.4โ๐จ EUVD-2026-116620 ๐ Score: 5.4/10 (CVSS v3.1) ๐ฆ Product: MLflow ๐ข Vendor: mlflow ๐ Updated: 2026-11-04 ๐ MLflow Trackin
A medium-severity vulnerability, EUVD-2026-116620, has been catalogued in MLflow, the open-source machine learning platform. The flaw, rated 5.4 out of 10 under CVSS v3.1, lies in the MLflow Tracking Server's jobs feature, which fails to enforce authorization. It allows remote attackers to disclose sensitive information. The advisory record was updated on November 4, 2026.
- 4New MLflow vulnerability allows remote privilege escalationโ๐จ EUVD-2026-116621 ๐ Score: 6.3/10 (CVSS v3.1) ๐ฆ Product: MLflow ๐ข Vendor: mlflow ๐ Updated: 2026-11-04 ๐ MLflow Trackin
A security flaw tracked as EUVD-2026-116621 has been catalogued in MLflow, the open-source machine learning platform. The bug, scored 6.3 out of 10 on the CVSS v3.1 scale, stems from missing authorization in the MLflow Tracking Server and lets remote attackers escalate privileges on affected installations. It was updated in the European vulnerability database on 4 November 2026. Administrators running MLflow are being urged to review their deployments and apply fixes.