@melaya/runner@1.1.27
Run Melaya AI pipelines locally with your own LM Studio or Ollama models
AI Security Review
scanned 2mo ago · by lpm-firewall-aiThe package has no install lifecycle hook, but its runtime design combines server-directed local code execution with capture and reuse of third-party authenticated browser sessions. This is a concrete dangerous capability requiring a warning rather than a clean classification.
Decision evidence
public snapshot- It downloads server-supplied bundle content with the runner token and writes it into a home-directory cache.
- It executes a Python pipeline after handling a server socket run command.
- It extracts LinkedIn and Luma authenticated session cookies and sends each login outcome over the socket.
- It persists the complete Luma browser storage state for later reuse.
- Its Luma bridge issues browser-context requests with included credentials, enabling use of the captured session.
- package.json has only a prepublish build hook, so installation does not automatically run the package.
- The observed login flows require an interactive browser sign-in and activation through a connected runner session.
- The package contains local model and browser-control functionality consistent with a remote pipeline runner, though the capability is high risk.
Source & flagged code
8 flaggedPackage source references child process execution.
dist/codeWorker.jsView on unpkg · L51Findings
Affected versions and remediation
This report applies to @melaya/runner@1.1.27.
See all versions of @melaya/runner for other recorded verdicts.
- Review the evidence and your use of @melaya/runner@1.1.27 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.