visionclaw@0.1.213
A personal assistant agent that runs on your desktop, receives commands from messaging channels, and executes tasks autonomously.
AI Security Review
scanned 2mo ago · by lpm-firewall-aiThis is not concrete malicious install-hook behavior, but it automatically changes a browser-agent extension surface and conditionally alters policy configuration. Downgrade to a warning for the guarded first-party extension lifecycle risk.
Decision evidence
public snapshot- The postinstall hook modifies the installed dependency tree.
- Runtime detection automatically configures a Chrome extension in a dedicated automation profile.
- The extension cleanup can invoke sudo to remove matching Chrome enterprise-policy entries.
- The postinstall code copies only a bundled local shim and makes no network request.
- The extension setup is limited to macOS and a package-owned automation profile.
- No runtime dependency on visionclaw itself is declared.
- Workspace API calls use credentials supplied in the workspace configuration and target Google APIs.
Source & flagged code
14 flaggedPackage defines install-time lifecycle scripts.
package.jsonView on unpkgPackage source references child process execution.
dist-agent/playwriter-relay.cjsView on unpkg · L6Findings
Affected versions and remediation
This report applies to visionclaw@0.1.213.
See all versions of visionclaw for other recorded verdicts.
- Review the evidence and your use of visionclaw@0.1.213 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.