@livx.cc/agentx@0.99.44
Edge-native AI agent runtime — drives a virtual filesystem via any LLM (ai.libx.js). Same bytes run in node, browser, or edge.
AI Security Review
scanned 1mo ago · by lpm-firewall-aiNo concrete credential theft, hidden exfiltration, or automatic preinstall/install/postinstall attack was found. The package nevertheless combines broad command execution and persistence-oriented agent tooling with a prepare-time repository mutation.
Decision evidence
public snapshot- The prepare script changes the consumer repository's Git hooks path.
- The CLI exposes agent tools that can create OS-scheduled jobs and load a macOS LaunchAgent.
- The agent Shell tool executes arbitrary commands through a system shell.
- A remote-trigger tool can resume another local agent session headlessly with automatic approval.
- There is no preinstall, install, or postinstall hook.
- The observed network credential request is conditional on user-supplied Bodify environment variables.
- No source shows secret harvesting or transmission to an undisclosed endpoint.
- The update check only queries the package's npm registry endpoint; installation requires the explicit update action.
Source & flagged code
6 flagged · loading sourcePackage source references a known benign dynamic code generation pattern.
dist/index.jsView on unpkg · L5357Manifest-reachable source sends a prompted API credential to a fixed unofficial gateway and persists the redirection.
dist/cli.jsView on unpkgFindings
Affected versions and remediation
This report applies to @livx.cc/agentx@0.99.44.
See all versions of @livx.cc/agentx for other recorded verdicts.
- Review the evidence and your use of @livx.cc/agentx@0.99.44 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.