Vulnerabilities in Training & MLOps
Experiment tracking, pipelines and compute clusters. 7 currently disclosed, newest first. Each links through to the merged record for that CVE.
- 7.5 MLflow vulnerable to arbitrary file read on unauthenticated servers via prompt model version source
- 6.5 MLflow CrewAI autologging records LLM provider API keys in logged trace spans
- 9.8 MLflow vulnerable to remote code execution on unauthenticated servers via third-party scorer deserialization importing caller-placed artifact modules
- 7.8 Hydra legacy target blacklist permits unsafe PyYAML loading
- 6.5 MLflow is an open source AI engineering platform for agents, large language models, and machine learning models.
- — MLflow Server-Side Request Forgery Vulnerability
- — Ray-Project Ray Code Injection Vulnerability
Open this in the live dashboard
to filter further, or query it with search_cves over MCP.