@agentvault/agentvault@0.23.53
End-to-end encrypted messaging for AI agents: connects an agent to its owner, rooms and other agents on AgentVault.
AI Security Review
scanned 11d ago · by lpm-firewall-aiInspected source does not support the scanner’s credential-redirection claim: the service defaults resolve to AgentVault, and setup is an explicit CLI flow. The package has no install hook, and the flagged command execution belongs to CLI diagnostics or setup.
Decision evidence
public snapshot- package.json has no install-time hook; its publish hook runs the build.
- The inspected runtime uses AgentVault’s service URL by default; the OpenClaw skill call uses the configured account URL or that same default.
- The CLI parses explicit arguments and documents setup as a command requiring an invite token.
- The dependency excerpts show WebSocket exports, browser behavior, and notification functionality consistent with their declared roles.
- The CLI includes shell and filesystem operations for diagnostics and setup, but the inspected code places them in CLI command flows rather than package installation.
- The OpenClaw integration sends skill invocation data and a device JWT to its configured AgentVault API endpoint.
Source & flagged code
10 flaggedSource sends credentials or rich application records to a package-controlled external receiver enabled by default.
dist/cli.jsView on unpkg · L129Findings
Affected versions and remediation
This report applies to @agentvault/agentvault@0.23.53.
See all versions of @agentvault/agentvault for other recorded verdicts.
- Review the evidence and your use of @agentvault/agentvault@0.23.53 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.