@livx.cc/agentx@0.99.46
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-aiThe unrestricted URL override creates a conditional credential exposure risk when the Bodify key is used. The source does not show covert activation or establish a malicious chain, so this warrants a warning rather than a block.
Decision evidence
public snapshot- When both Bodify credentials are set, the CLI sends the bearer key to the URL selected by `BODIFY_URL` without destination validation.
- The CLI calls this credential-fetching path when its main command runs.
- The cited CLI location constructs `.claude/worktrees` paths for Git worktrees.
- The agent type declaration describes VFS-backed agent definitions, and the library loads project instruction files through its supplied filesystem.
- The manifest’s `prepare` hook configures a Git hook path; it defines no install lifecycle hook.
- The credential request requires both `BODIFY_API_KEY` and `BODIFY_APP_ID`, and defaults to `bodify.bod.ee` when no URL override is set.
- The inspected agent integration excerpts describe worktree paths, VFS-backed definitions, and project instruction loading; they do not establish an agent control-surface attack.
Source & flagged code
6 flagged · loading sourcePackage source references a known benign dynamic code generation pattern.
dist/index.jsView on unpkg · L5409Manifest-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@0.99.46.
See all versions of @livx.cc/agentx for other recorded verdicts.
- Review the evidence and your use of @livx.cc/agentx@0.99.46 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.