Evidence · DiggingBeagle record
MLflow FileStore model_id permitted metadata writes outside the run directory (original advisory or patch)
An MLflow tracking API used an untrusted model identifier as a filesystem path component in FileStore, allowing out-of-run metadata writes if the service was reachable.
- Published
- Oct 26, 2026
- Source role
- patch commit artifact
Evidence record
An MLflow tracking API used an untrusted model identifier as a filesystem path component in FileStore, allowing out-of-run metadata writes if the service was reachable.
Claim-level citations (4)
- supportsMLflow FileStore model_id permitted metadata writes outside the run directory: The MLflow pull request includes regression checks for model identifier path handling. Its reported pre-fix run failed six targeted checks and the patched suite passed 31 checks, supplying maintainer-side validation of FileStore confinement rather than independent evidence about deployed clients or exploitation.
PR #27131 test description and regression output for FileStore model_id validation, merged October 2026
- supportsMLflow FileStore model_id permitted metadata writes outside the run directory: PR 27131 merged in October and rejects absolute and parent-relative model identifiers in the FileStore path; no downstream model-loading exploit or actual affected customer has been independently demonstrated.
PR 27131 summary and validation-test changes
- supportsMLflow FileStore model_id permitted metadata writes outside the run directory: The FileStore log_outputs implementation joined caller-supplied model_id into a path without requiring a single safe path component; the fix adds component validation and traversal tests.
Implementation diff, validate model id as a single path component
- supportsMLflow FileStore model_id permitted metadata writes outside the run directory: MLflow PR 27131 added validation that log_outputs model IDs are single path components, rejecting absolute paths and parent-relative or Windows-normalized traversal; the defect is specific to the FileStore backend and does not itself demonstrate arbitrary code execution.
MLflow pull request 27131, file_store.py log_outputs validation and regression tests