Lucene search
+L

3 matches found

OSV
OSV
•added 2026/09/15 8:48 p.m.•11 views

GHSA-5648-RGJ9-V224 @zereight/mcp-gitlab has multiple safety-control bypasses: execute_graphql read-only + allow-list bypass, unauthenticated transports, session-exhaustion DoS

Summary @zereight/mcp-gitlab exposes GitLab to an LLM agent while relying on read-only mode, a project allow-list, and transport auth as its safety controls. Five defects defeat those controls. Under the MCP threat model, tool-call arguments/content can be shaped by untrusted input prompt injecti...

8.1CVSS6AI score
SaveExploits0References8
OSV
OSV
•added 2026/09/01 5:04 p.m.•6 views

GHSA-GQVG-GMMX-X4HM MLFLOW_ALLOW_PICKLE_DESERIALIZATION=False safety control bypassed by mlflow.statsmodels flavor — RCE via crafted model artifact

Summary MLflow introduced MLFLOWALLOWPICKLEDESERIALIZATION as a security control to prevent unsafe pickle.load execution during model loading, in response to CVE-2024-37052 through CVE-2024-37060. When set to False, operators expect all pickle deserialization to be blocked. The most recent relate...

8.8CVSS6.3AI score
SaveExploits0References5
Github Security Blog
Github Security Blog
•added 2026/09/01 5:04 p.m.•30 views

MLFLOW_ALLOW_PICKLE_DESERIALIZATION=False safety control bypassed by mlflow.statsmodels flavor — RCE via crafted model artifact

Summary MLflow introduced MLFLOWALLOWPICKLEDESERIALIZATION as a security control to prevent unsafe pickle.load execution during model loading, in response to CVE-2024-37052 through CVE-2024-37060. When set to False, operators expect all pickle deserialization to be blocked. The most recent relate...

6.3AI score
SaveExploits0References5Affected Software1
Rows per page
Query Builder