kixparadigm@1.3.7
kixparadigm — AI 自编排最小范式(认知层常驻)+ kixpower 多智能体编排 + DSH 一键导入(preset + vision-bridge + 门禁插件)
AI Security Review
scanned 1mo ago · by lpm-firewall-aiThis is an automatic first-party AI-agent extension installation with meaningful code-execution capability, so it warrants a warning. The inspected source does not show a concrete malicious payload, covert credential theft, or unrelated network recipient.
Decision evidence
public snapshot- The package runs its installer automatically through a postinstall hook.
- The installer copies preset content into the DSH agent-preset directory and adds a plugin mount to the DSH web profile.
- The installed preset enables shell tools and registers a probe tool that runs caller-provided Python in the active workspace.
- The vision component reads a named DSH credential and sends image-description requests only to a fixed BigModel endpoint.
- No hidden endpoint, encoded payload, self-dependency, or unrelated credential recipient was found in the inspected active paths.
- The credential is used for the named vision service endpoint rather than an operator-configured destination.
Source & flagged code
20 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.
dsh/preset-classic/plugins/kix-guards.jsView on unpkg · L152Findings
Affected versions and remediation
This report applies to kixparadigm@1.3.7.
See all versions of kixparadigm for other recorded verdicts.
- Review the evidence and your use of kixparadigm@1.3.7 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.