@mulmoclaude/core@4.4.2
Shared server-side core for MulmoClaude and MulmoTerminal — the always-shipped-together subsystems consolidated behind subpath exports so the two hosts can't drift. Server-only except the browser-safe ./artifacts, ./whisper/client, ./workspace-setup/slug,
AI Security Review
scanned 2mo ago · by lpm-firewall-aiThe package contains a real project-scoped Claude skill installation capability, so it warrants a warning under the agent-extension lifecycle policy. Source inspection found no install hook or concrete malicious behavior.
Decision evidence
public snapshot- Runtime skill bridge mirrors allowlisted staged skill files into <workspace>/.claude/skills/<slug>.
- Registry import can download a collection bundle and, when configured, mirror it into the Claude skills directory.
- This is an agent-extension lifecycle capability, not install-time malware.
- package.json has no preinstall, install, postinstall, or other lifecycle hook.
- The bridge restricts slugs and file paths to an allowlist; no broad Claude configuration write was found.
- No credential harvesting, environment exfiltration, eval/vm execution, or covert network beacon was found.
Source & flagged code
4 flagged · loading sourcePackage metadata claims a different repository identity while copied source loads a runtime dependency bridge.
dist/discovery-P64fgtuD.jsView on unpkg · L1436Findings
Affected versions and remediation
This report applies to @mulmoclaude/core@4.4.2.
See all versions of @mulmoclaude/core for other recorded verdicts.
- Review the evidence and your use of @mulmoclaude/core@4.4.2 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.