@livx.cc/agentx@0.99.115
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 flagged agent-related locations support project-scoped worktree and instruction utilities, while runtime subagent definitions are loaded from a caller-configured directory. The manifest's prepare hook configures Git hooks, and the Bodify credential is used for the matching environment retrieval feature; the inspected source did not establish an attack.
Decision evidence
public snapshot- The manifest exposes the runtime and CLI; its prepare hook only configures the repository's Git hooks path.
- The CLI uses a Bodify token to request environment values from the Bodify endpoint and loads returned values into its process environment.
- The agent integration locations describe project worktrees, project instruction loading, and typed agent definitions; runtime agent files are loaded only when an agents directory is configured.
- The shell tool filters common secret environment variables from child processes by default, and its declarations describe filesystem and network sandbox limits.
- The MCP client exports HTTP and stdio transports as runtime API capabilities.
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@0.99.115.
See all versions of @livx.cc/agentx for other recorded verdicts.
- Review the evidence and your use of @livx.cc/agentx@0.99.115 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.