@livx.cc/agentx@0.99.29
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 14d ago · by lpm-firewall-aiInspected source shows no install-time attack: the prepare hook only configures a Git hooks path. The credential and agent integration findings are consistent with caller-configured Bodify access and project-local agent features, so the package is clean.
Decision evidence
public snapshot- package.json defines only a prepare hook that configures a Git hooks path; it has no preinstall, install, or postinstall hook.
- The CLI worktree code uses a .claude/worktrees directory; this is not an agent configuration write.
- The type declaration describes subagent definitions loaded from files through the package filesystem abstraction.
- The library reads project instruction files from its virtual filesystem; this code does not persist or modify agent settings.
- Bodify credential retrieval requires both caller-provided Bodify credentials and requests secrets from the Bodify endpoint.
- CLI startup loads Bodify secrets before collecting provider keys; it does not prompt for and redirect an existing provider key.
Source & flagged code
6 flagged · loading sourcePackage source references a known benign dynamic code generation pattern.
dist/index.jsView on unpkg · L4863Manifest-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@0.99.29.
See all versions of @livx.cc/agentx for other recorded verdicts.
- Review the evidence and your use of @livx.cc/agentx@0.99.29 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.