@vibes.diy/api-svc@14.3.273
WebSocket-based API for vibes.diy chat persistence and app deployment. Supports both HTTP POST and WebSocket connections using the same handler via the Evento event-driven framework.
AI Security Review
scanned 3h ago · by lpm-firewall-aiThe flagged avatar script is a dry-run-by-default backfill that fetches avatar assets from the package's host and requires --apply for database and object-storage writes. Embeddings use an explicitly configured backend, while alarm jobs use GitHub credentials for repository issue handling in write mode; no malicious behavior was identified.
Decision evidence
public snapshot- The avatar backfill fetches avatar data using handle and asset URL parameters; its default base is https://vibes.diy.
- Avatar backfill uses a read connection by default and requires --apply to select the admin database connection.
- Embedding backfill requires --apply for writes and requires an explicitly configured LLM backend URL and API key.
- Embedding text is sent with the configured backend API key to the configured backend's embeddings endpoint.
- The alarm uses GitHub credentials for its configured repository, and issue operations are gated by its write mode.
Source & flagged code
5 flaggedSource sends credentials or rich application records to a package-controlled external receiver enabled by default.
usage-report/backfill-avatars.jsView on unpkg · L1The avatar backfill fetches avatar data using handle and asset URL parameters; its default base is https://vibes.diy.
usage-report/backfill-avatars.jsView on unpkg · L30Findings
Affected versions and remediation
This report applies to @vibes.diy/api-svc@14.3.273.
See all versions of @vibes.diy/api-svc for other recorded verdicts.
- Review the evidence and your use of @vibes.diy/api-svc@14.3.273 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.