@clawos-dev/clawd@0.2.554
Standalone clawd daemon — Claude Code (and future Codex) session server over WebSocket
AI Security Review
scanned 1mo ago · by lpm-firewall-aiThe package does not automatically run at npm install, but its active CLI exposes an owner-triggered, package-controlled remote shell bootstrap without integrity verification. That is a concrete dangerous capability, not sufficient evidence of automatic malware behavior.
Decision evidence
public snapshot- The CLI fetches a setup script from a package-controlled portal and executes its returned text with Bash.
- The execution environment inherits the daemon environment and adds the caller-supplied pool key.
- The owner-authorized pool-connection RPC invokes that remote-script path after accepting an API key.
- This is a remotely mutable code path with no source-visible integrity check on the fetched script.
- package.json has no preinstall, install, or postinstall lifecycle hook.
- The remote-script path requires the owner-only engine:connectPool RPC and an API key supplied in that request.
Source & flagged code
19 flagged · loading sourcePackage source references child process execution.
dist/app-builder-plugin/mcp-server.cjsView on unpkg · L21117Source file is highly similar to a previously finalized malicious package; route for source-aware review.
dist/app-builder-plugin/mcp-server.cjsView on unpkgPackage contains a possible secret pattern.
dist/app-builder-plugin/mcp-server.cjsView on unpkg · L24903Findings
Affected versions and remediation
This report applies to @clawos-dev/clawd@0.2.554.
See all versions of @clawos-dev/clawd for other recorded verdicts.
- Review the evidence and your use of @clawos-dev/clawd@0.2.554 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.