@livx.cc/agentx@0.99.95
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 scanner's credential-redirection and persistence hints are not supported by the inspected source. The observed vault lookup uses a Bodify-specific token and endpoint for provider configuration; no concrete attack behavior was established.
Decision evidence
public snapshot- package.json defines prepare as a local Git hook-path setting and prepublishOnly as a build command.
- The CLI uses a Bodify vault token only when available, then requests app environment values from the Bodify service; startup calls this lookup to load provider keys.
- The agent integration locations describe Claude worktree paths, agent definitions, and reading project instruction files; they do not establish a configuration write.
- The shell and MCP client sources describe configured tools and a sandboxed shell capability; the inspected excerpts show no package-initiated attack.
- The CLI can send a Bodify vault token to the configured Bodify URL to retrieve environment values.
- The package includes shell command execution capability.
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@0.99.95.
See all versions of @livx.cc/agentx for other recorded verdicts.
- Review the evidence and your use of @livx.cc/agentx@0.99.95 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.