karajan-code@4.31.0
Local multi-agent coding orchestrator with TDD, SonarQube, and code review pipeline
AI Security Review
scanned 23d ago · by lpm-firewall-aiSource proves automatic first-party MCP setup without a consent prompt. The supplied policy classifies package-owned agent extension setup as a warning unless separate concrete attack behavior is established.
Decision evidence
public snapshot- package.json automatically runs scripts/postinstall.js after installation.
- The postinstall caller configures both agents without a consent gate.
- The Claude configuration receives a karajan-mcp entry launching the package's local server.
- The Codex configuration is updated with the package's MCP registration.
- The registration targets the package's own MCP server rather than an unrelated payload.
- The inspected hook preserves unrelated configuration entries; no credential forwarding or remote payload execution was identified.
Source & flagged code
24 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-code@4.31.0.
See all versions of karajan-code for other recorded verdicts.
- Review the evidence and your use of karajan-code@4.31.0 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.