@livx.cc/agentx@0.99.93
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 inspected source supports ordinary agent runtime behavior and does not establish the scanner’s credential-redirection or persistence claims. The prepare hook configures a Git hooks path, and the flagged agent locations show reads or runtime declarations rather than an unconsented control-surface write.
Decision evidence
public snapshot- package.json declares only build, test, and prepare/publish hooks; prepare sets the Git hooks path to `.githooks`.
- The flagged CLI location handles Git worktree paths; Claude permission settings are read and merged from settings files.
- The flagged declaration describes VFS-backed agent definitions; the runtime loads project instruction files through its filesystem interface.
- Shell execution is exposed as a runtime tool, and MCP client process support is part of the package runtime.
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@0.99.93.
See all versions of @livx.cc/agentx for other recorded verdicts.
- Review the evidence and your use of @livx.cc/agentx@0.99.93 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.