@mulmoclaude/core@0.23.0
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 ./whisper/client, ./workspace-setup/slug, ./translatio
AI Security Review
scanned 3mo ago · by lpm-firewall-aiThe package is not malicious by source evidence, but its explicit runtime import capability writes remote skill content into an AI-agent control surface. Policy classifies this guarded first-party extension setup as warn-worthy.
Decision evidence
public snapshot- Runtime registry import downloads collection bundles and mirrors allowlisted files into `.claude/skills`.
- Imported `SKILL.md`, schema, and templates can affect Claude skill discovery.
- Whisper helper downloads large models and invokes configured local `ffmpeg`/`whisper-server` binaries.
- `package.json` has no preinstall, install, or postinstall hook.
- Skill mirroring is runtime-only, path/slug allowlisted, and scoped under the workspace.
- Google OAuth/token handling is explicit API functionality with HTTPS and consent-host checks.
- No hidden exfiltration, eval, shell execution, or import-time network activity found.
Source & flagged code
1 flagged · loading sourceThis package version adds a dangerous source file absent from the previous stored version; route for source-aware review.
dist/google/index.cjsView on unpkgFindings
Affected versions and remediation
This report applies to @mulmoclaude/core@0.23.0.
See all versions of @mulmoclaude/core for other recorded verdicts.
- Review the evidence and your use of @mulmoclaude/core@0.23.0 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.