logod-chrome-mcp@1.0.5
로고드 크롬 엠시피 — 클로드가 내 크롬을 직접 조종하게 해주는 전용 도구
AI Security Review
scanned 7d ago · by lpm-firewall-aiThe postinstall hook automatically installs a package-owned Claude skill that can guide later agent-run commands, so a warning is warranted. The inspected behavior does not establish a concrete attack supporting a block.
Decision evidence
public snapshot- The package postinstall hook runs during installation; global installs call the Claude skill setup automatically.
- The setup writes a package-owned skill under the user’s Claude skills directory when that directory exists; an environment variable can opt out.
- The installed skill directs Claude to run the package’s diagnostic and repair commands through Bash when users report related problems.
- The setup is scoped to a named package skill and skips installation when the Claude directory is absent.
- The inspected source establishes agent configuration setup but does not establish credential theft or another concrete attack.
Source & flagged code
16 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 unpkgFindings
Affected versions and remediation
This report applies to logod-chrome-mcp@1.0.5.
See all versions of logod-chrome-mcp for other recorded verdicts.
- Review the evidence and your use of logod-chrome-mcp@1.0.5 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.