@chem-x/starter-kit@26.9.20-1318
Chemical X Protocol: Private drop-in architecture starter kit and capsule generator
AI Security Review
scanned 11d ago · by lpm-firewall-aiThe package’s automatic npm postinstall writes an MCP server entry into consumer project agent configuration without an explicit opt-in. This meets the policy for blocking unconsented install-time mutation of an AI-agent control surface.
Decision evidence
public snapshot- The manifest runs cli/postinstall.js automatically after installation.
- The postinstall handler calls installAllMcpConfigs for the installing project without a consent check.
- The MCP installer writes a chemical-x server entry into the project's Cursor MCP configuration.
- The installer also writes MCP configuration for VS Code; its test verifies the Cursor entry is created.
- A separate wizard writes agent guidance only when pillars are selected.
- The installer preserves existing MCP server entries while adding its own entry.
- The pillars wizard’s agent-file writes are conditional on selected pillars and are separate from the postinstall path.
Source & flagged code
10 flagged · loading sourcePackage defines install-time lifecycle scripts.
package.jsonView on unpkgInstall-time lifecycle script is not statically allowlisted and needs review.
package.jsonView on unpkgPackage contains a critical-looking secret pattern.
cli/errors/errors.spec.jsView on unpkg · L27Findings
Affected versions and remediation
This report applies to @chem-x/starter-kit@26.9.20-1318.
See all versions of @chem-x/starter-kit for other recorded verdicts.
- Avoid installing @chem-x/starter-kit@26.9.20-1318. Remove it from direct dependencies and check your lockfile for transitive copies.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
- If this version ran, investigate the affected machine and build environment. Rotate credentials it could access and rebuild from a trusted environment.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.