create-chemx@26.9.12-1140
Chemical X Protocol: Private drop-in architecture starter kit and capsule generator
AI Security Review
scanned 1mo ago · by lpm-firewall-aiThe package has no install lifecycle execution, but its user-invoked scaffold downloads server-controlled content and writes unchecked paths to disk. This creates a concrete remote payload and path-traversal capability that warrants a warning.
Decision evidence
public snapshot- The scaffold fetches server-provided starter-kit content after sending a licence key and persistent device identifier.
- Returned file paths are joined and written without validating that they remain inside the chosen project directory.
- The scaffold also writes an AI-editor rules file into the created project.
- package.json has no install lifecycle hook or self-dependency.
- The risky download-and-write path is reached only by an explicit create or init CLI command after licence input.
- No source path executes the downloaded content during scaffolding.
Source & flagged code
5 flagged · loading sourcePackage source references child process execution.
cli/navigator-share.jsView on unpkg · L2A single source file combines environment access, network access, and code or shell execution; review context before blocking.
cli/license.jsView on unpkg · L3This package version adds a dangerous source file absent from the previous stored version; route for source-aware review.
cli/license.jsView on unpkgManifest-trigger-reachable source writes behavior-bearing configuration into a user or project AI-agent control surface.
cli/scaffold.jsView on unpkgFindings
Affected versions and remediation
This report applies to create-chemx@26.9.12-1140.
See all versions of create-chemx for other recorded verdicts.
- Review the evidence and your use of create-chemx@26.9.12-1140 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.