@octavus/agent@1.0.16
Octavus Agent CLI - run one of your cloud Octavus Agents on a machine you control, using that machine as the agent's computer. The agent lives in the cloud (fully configured and observable); the local machine only supplies the computer and the trigger.
AI Security Review
scanned 29d ago · by lpm-firewall-aiThe configurable destination and token forwarding create a concrete conditional credential-exposure risk. The package has no lifecycle hook or evidence of an automatic malicious action.
Decision evidence
public snapshot- The CLI accepts an operator WebSocket URL from a command value, environment variable, or saved configuration without a host allowlist.
- It builds a WebSocket endpoint from that URL and sends the computer token as a query parameter.
- Package metadata exposes a CLI entry point and contains no npm lifecycle hook.
- No install-time script or self-dependency is declared.
- No hidden default recipient or confirmed unauthorized data transfer was identified.
Source & flagged code
12 flagged · loading sourcePackage source references child process execution.
dist/computer-mcp/dist/index.jsView on unpkg · L53Source file is highly similar to a previously finalized malicious package; route for source-aware review.
dist/computer-mcp/dist/index.jsView on unpkgFindings
Affected versions and remediation
This report applies to @octavus/agent@1.0.16.
See all versions of @octavus/agent for other recorded verdicts.
- Review the evidence and your use of @octavus/agent@1.0.16 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.