@karajan-family/code@4.52.0
Local multi-agent coding orchestrator with TDD, SonarQube, and code review pipeline
AI Security Review
scanned 7d ago · by lpm-firewall-aiSource confirms unprompted, persistent first-party MCP registration by an install hook, which warrants a warning for agent extension lifecycle risk. The inspected behavior does not establish a concrete attack that supports blocking publication.
Decision evidence
public snapshot- package.json runs scripts/postinstall.js automatically; its main routine invokes both agent registration functions.
- The hook writes a karajan-mcp entry into the home Claude and Codex configurations that launches this package's local MCP server over stdio.
- The other inspected integration targets concern separate scoped behavior, including an explicit kj-trash install command, project setup, checks, file filtering, and guideline destinations.
- The registration points to this package's local server; the inspected hook does not show credential collection, network transfer, or unrelated permission grants.
- The ai-trash hook installer is reached through an explicit install command that requires the --claude-code flag.
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@4.52.0.
See all versions of @karajan-family/code for other recorded verdicts.
- Review the evidence and your use of @karajan-family/code@4.52.0 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.