@mulmoclaude/core@4.9.0
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 1mo ago · by lpm-firewall-aiThe package has a concrete runtime path that installs remotely obtained skill content into a project AI-agent control surface. It lacks install hooks, self-dependency, obfuscation, credential exfiltration, or an unconsented automatic mutation chain, so the appropriate result is a warning rather than a block.
Decision evidence
public snapshot- A runtime collection import downloads a selected bundle and mirrors allowlisted files into a Claude skill directory.
- The skill bridge writes staged skill content into the project .claude/skills control surface.
- The package contains runtime network clients for registry imports, feeds, OAuth, and model downloads.
- package.json has no preinstall, install, postinstall, or self-dependency.
- The skill bridge restricts slugs and mirrored file paths to an allowlist.
- Feed retrieval rejects non-HTTP URLs and private or loopback destinations.
- OAuth broker calls require HTTPS except for loopback testing.
Source & flagged code
4 flagged · loading sourcePackage metadata claims a different repository identity while copied source loads a runtime dependency bridge.
dist/discovery-2TvulIz8.cjsView on unpkg · L1451Findings
Affected versions and remediation
This report applies to @mulmoclaude/core@4.9.0.
See all versions of @mulmoclaude/core for other recorded verdicts.
- Review the evidence and your use of @mulmoclaude/core@4.9.0 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.