karajan-code@4.49.5
Local multi-agent coding orchestrator with TDD, SonarQube, and code review pipeline
AI Security Review
scanned 10d ago · by lpm-firewall-aiThe postinstall hook performs unprompted first-party MCP setup in two agent configuration files. The source does not establish a broader control-surface write or concrete malicious behavior, so this warrants a warning rather than a block.
Decision evidence
public snapshot- package.json runs scripts/postinstall.js as an npm postinstall hook.
- The hook directly registers this package's local karajan-mcp server in Claude configuration and writes the configuration without a consent prompt.
- The same hook adds a karajan-mcp server block to Codex configuration and writes the file.
- The separate ai-trash hook installer requires the explicit kj-trash install --claude-code command.
- Other inspected integration locations concern project skill installation, configuration checks, file filtering, or agent guideline targets.
- The automatic configuration change is limited to this package's named MCP server and local server path; inspected source shows no credential transfer or remote endpoint in this setup.
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@4.49.5.
See all versions of karajan-code for other recorded verdicts.
- Review the evidence and your use of karajan-code@4.49.5 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.