agenticros@0.7.34
AgenticROS - agentic AI for ROS-powered robots. Single CLI to launch real-robot or simulated demos, manage configuration, and inspect status.
AI Security Review
scanned 12d ago · by lpm-firewall-aiInspected source shows guarded first-party agent setup, including an Antigravity permission grant for the package's own MCP tools. This supports a warning for agent extension lifecycle risk, but not a malicious verdict.
Decision evidence
public snapshot- The package has no install hook; its only lifecycle hook is prepublishOnly, which runs build and packaging checks.
- The setup wizard asks before installing the OpenClaw plugin and runs its installer only after consent.
- The wizard separately asks before configuring MCP clients and invokes setup only on approval.
- Antigravity setup adds an allow rule covering every tool exposed by the AgenticROS MCP server.
- The agent configuration changes are initiated through the package setup flow, which prompts the user before applying them.
- The Antigravity permission rule is limited to the package's own MCP server; other tools remain subject to their existing permissions.
Source & flagged code
24 flagged · loading sourcePackage source references child process execution.
dist/util/mcp-discovery.jsView on unpkg · L16Source file is highly similar to a previously finalized malicious package; route for source-aware review.
dist/util/mcp-discovery.jsView on unpkgFindings
Affected versions and remediation
This report applies to agenticros@0.7.34.
See all versions of agenticros for other recorded verdicts.
- Review the evidence and your use of agenticros@0.7.34 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.