@spuentesp/lain-mcp@0.8.0
High-performance knowledge graph MCP server for AI coding agents with semantic search.
AI Security Review
scanned 15d ago · by lpm-firewall-aiThe postinstall installs the package’s native component from its upstream release host, but the checksum is fetched from that same host and does not independently authenticate the binary. This is a real supply-chain risk without inspected evidence of malicious behavior, so warn rather than block.
Decision evidence
public snapshot- The package’s postinstall calls an installer that downloads a native release archive and checksum from GitHub.
- The archive is checked against the downloaded checksum, but the checksum and archive come from the same release host.
- The installer extracts and runs the binary for a version check, then stores it in a versioned cache.
- The download targets the package’s declared upstream repository and a release matching the package version by default.
- The postinstall skips the download in CI unless explicitly forced, and when npm scripts are disabled.
- The source shows no credential collection, unrelated network destination, or mutation of an AI-agent control surface.
Source & flagged code
5 flagged · loading sourcePackage defines install-time lifecycle scripts.
package.jsonView on unpkgInstall-time source downloads a native archive from a fixed external host without transport verification, extracts it, and installs an executable payload.
package.jsonView on unpkg · L2Findings
Affected versions and remediation
This report applies to @spuentesp/lain-mcp@0.8.0.
See all versions of @spuentesp/lain-mcp for other recorded verdicts.
- Review the evidence and your use of @spuentesp/lain-mcp@0.8.0 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.