@karajan-family/code@4.44.0
Local multi-agent coding orchestrator with TDD, SonarQube, and code review pipeline
AI Security Review
scanned 9d ago · by lpm-firewall-aiThe install hook makes a persistent, unprompted change to global agent configuration by registering this package’s own server. This fits the warning policy for first-party agent extension setup; the inspected source does not establish a concrete malicious action.
Decision evidence
public snapshot- The manifest runs the postinstall script during npm installation, and that script proceeds without a consent prompt.
- The hook adds a package-owned MCP server to the user’s global Claude and Codex configuration, launching this package’s local server through stdio.
- The separate trash-hook installer requires an explicit install command and flag before changing Claude settings.
- The init command asks for confirmation before installing slash-command files; the other inspected flagged paths show project hygiene, configuration checks, file watching, agent-config inspection, or legacy-block cleanup.
- The registration code inspected here does not show a remote endpoint or a content-export setting.
Source & flagged code
25 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 source references child process execution.
scripts/dual-publish.mjsView on unpkg · L17Findings
Affected versions and remediation
This report applies to @karajan-family/code@4.44.0.
See all versions of @karajan-family/code for other recorded verdicts.
- Review the evidence and your use of @karajan-family/code@4.44.0 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.