@livx.cc/agentx@0.99.110
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 4d ago · by lpm-firewall-aiThe manifest's prepare hook configures local Git hooks, and the inspected runtime behavior is consistent with the package's CLI, agent, shell, and MCP features. The credential requests inspected use configured or package-aligned destinations, with no confirmed attack behavior.
Decision evidence
public snapshot- package.json defines prepare to configure this repository's Git hooks and prepublishOnly to build; it has no install hook.
- The CLI loads Bodify credentials from its environment or macOS keychain and requests app secrets from bodify.bod.ee; startup uses the returned values as provider API keys.
- The flagged agent-related code describes file-based child-agent definitions and loads project instruction files relative to the current working directory.
- The shell module runs commands through its exposed Shell tool and supports an optional OS sandbox.
- The MCP client sends its configured bearer token to its configured server URL.
Source & flagged code
12 flagged · loading sourcePackage source references child process execution.
dist/tools.shell.jsView on unpkg · L891Findings
Affected versions and remediation
This report applies to @livx.cc/agentx@0.99.110.
See all versions of @livx.cc/agentx for other recorded verdicts.
- Review the evidence and your use of @livx.cc/agentx@0.99.110 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.