channel-worker@2.5.101
Package no longer supported. Contact Support at https://www.npmjs.com/support for more info.
Channel Manager worker daemon — runs on remote machines to execute video pipeline jobs
AI Security Review
scanned 8d ago · by lpm-firewall-aiThe cache route forwards worker-authenticated data to unauthenticated local callers and allows cross-origin reads. This is a real security risk, but the inspected source does not establish active malicious behavior or the scope of data exposed, so a warning is appropriate.
Decision evidence
public snapshot- The CLI starts the daemon, which starts the local cache server.
- The cache server binds to 127.0.0.1:8849 and listens for requests there.
- A GET to /cache/s3 accepts a file key without an authentication check and returns the resolved data.
- The response allows reads from any web origin with wildcard CORS.
- The cache server uses the worker token to fetch the requested data from the dashboard API.
- The server binds to loopback, limiting direct access to the local machine.
Source & flagged code
5 flaggedPackage source references dynamic require/import behavior.
bin/cli.jsView on unpkg · L2Package source invokes a package manager install command at runtime.
lib/updater.jsView on unpkg · L29A bounded semantic-analysis stage reached its safety limit; remaining detectors completed, but this package requires AI review.
lib/daemon.js#virtual:normalized:round1View on unpkgThis package version adds a dangerous source file absent from the previous stored version; route for source-aware review.
scripts/upload_facebook.jsView on unpkgFindings
Affected versions and remediation
This report applies to channel-worker@2.5.101.
See all versions of channel-worker for other recorded verdicts.
- Review the evidence and your use of channel-worker@2.5.101 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.