AI Security Review
scanned 2h ago · by lpm-firewall-aiReview flagged AI-agent configuration or capability changes. This remains warn-only unless evidence shows foreign-agent hijack through preinstall/install/postinstall, hidden persistence, exfiltration, remote code execution, or other concrete malicious behavior.
Decision evidence
public snapshot- dist/index.js exposes agent tools that read, write, and execute user-project commands.
- dist/index.js /google auth scans download folders for client_secret_*.json after explicit command.
- dist/index.js supports user-command MCP server configuration with arbitrary stdio command.
- dist/index.js sends configured chat/OAuth requests to third-party provider APIs.
- package.json preinstall only checks the Node major version.
- package.json postinstall only chmods dist/index.js.
- No install hook invokes network, credential collection, or foreign agent-config mutation.
- Google credential import and OAuth flows are reachable through explicit /google commands.
- No concrete source path sends harvested local credentials to an unrelated endpoint.
Source & flagged code
8 flagged · loading sourceInstall-time lifecycle script matches a deterministic static-gate block pattern.
package.jsonView on unpkgPackage defines install-time lifecycle scripts.
package.jsonView on unpkgInstall-time lifecycle script is not statically allowlisted and needs review.
package.jsonView on unpkgA single source file combines environment access, network access, and code or shell execution with blocking evidence.
dist/index.jsView on unpkg · L14A package entrypoint or install-time lifecycle script reaches a source file with blocking dangerous behavior.
dist/index.jsView on unpkg · L14Source combines command execution, command-output handling, and outbound requests; review data flow before blocking.
dist/index.jsView on unpkg · L14Package source references dynamic require/import behavior.
dist/index.jsView on unpkg · L2