@agentvault/agentvault@0.23.50
End-to-end encrypted messaging for AI agents: connects an agent to its owner, rooms and other agents on AgentVault.
AI Security Review
scanned 13d ago · by lpm-firewall-aiSource inspection found no concrete attack behavior. Scanner claims were not supported by the inspected endpoint, telemetry, CLI, plugin, or dependency code.
Decision evidence
public snapshot- package.json exposes the CLI and plugin entrypoints; its only publish lifecycle is prepublishOnly, which runs the build.
- The plugin defaults to AgentVault’s API, and setup accepts an endpoint supplied by the caller.
- Telemetry uses the configured API base and sends span metadata such as tool names; inspected fields do not include prompt or tool content.
- The flagged WebSocket dependency excerpts handle protocol headers, while the notifier excerpts select platform notifications and filter option flags.
- The dynamic require in the plugin resolves OpenClaw to check its version.
- The inspected source does not substantiate the scanner’s claim of a fixed unofficial credential gateway.
- The credentialed telemetry request targets the configured API base; the default is https://api.agentvault.chat.
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 · L120Findings
Affected versions and remediation
This report applies to @agentvault/agentvault@0.23.50.
See all versions of @agentvault/agentvault for other recorded verdicts.
- Review the evidence and your use of @agentvault/agentvault@0.23.50 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.