@melaya/runner@1.1.74
Run Melaya AI pipelines locally with your own LM Studio or Ollama models
AI Security Review
scanned 5h ago · by lpm-firewall-aiThe flagged credential flows use provider credentials with the matching provider APIs to discover available models, and the browser login flow is request-triggered. The package has no install hook; its limited failure telemetry requires a runner token and targets Melaya.
Decision evidence
public snapshot- The only lifecycle hook is prepublishOnly, which runs the build.
- The CLI starts provider discovery after launch; provider credentials are used to query their corresponding services for model availability.
- Claude credentials are sent to api.anthropic.com, and GitHub Copilot credentials are exchanged with api.github.com before querying api.githubcopilot.com.
- Install-failure reporting requires a runner token, sends bounded stage and platform details to the configured Melaya server, and supports an opt-out.
- Luma browser login is triggered by a server login request, not automatically at import or startup.
- Browser authorization code classifies cloud metadata and private addresses as forbidden.
- The static scanner flags process execution, credential flows, and cloud metadata references, but inspected instances serve browser setup, provider discovery, or address blocking.
- The manifest has no preinstall, install, or postinstall hook.
- The package describes connecting local models to Melaya pipelines, consistent with provider discovery and the inspected service calls.
Source & flagged code
20 flaggedPackage source references child process execution.
dist/uvBootstrap.jsView on unpkg · L20Source file is highly similar to a previously finalized malicious package; route for source-aware review.
dist/uvBootstrap.jsView on unpkgFindings
Affected versions and remediation
This report applies to @melaya/runner@1.1.74.
See all versions of @melaya/runner for other recorded verdicts.
- Review the evidence and your use of @melaya/runner@1.1.74 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.