grix-connector
Connect local AI coding agents (Claude, Codex, Gemini, Qwen, DeepSeek, Cursor, OpenCode, Pi, OpenHuman, Reasonix) to the Grix scheduling platform. Also serves as an OpenClaw plugin for Grix channel transport.
This report shows the evidence for grix-connector@4.11.1. 3 of 86 known versions are blocked or flagged. The newest version, 4.16.1, is allowed. See all versions.
AI Security Review
scanned 1mo ago · by lpm-firewall-aiThe package contains an automatic install-time persistence action and a bundled skill that directs agents to install arbitrary remote skill packages over existing skills. The latter establishes a concrete remote payload path rather than ordinary connector configuration.
Decision evidence
public snapshot- The install hook automatically writes executable guardian files into the user's Grix home directory.
- A bundled agent skill directs an agent to download, unpack, and overwrite skills from a supplied URL.
- The skill treats a platform instruction as sufficient authority for self-installation, enabling remote payload delivery into agent skill directories.
- No runtime self-dependency on grix-connector is declared.
- The install hook's direct writes are limited to the package's .grix directory.
Source & flagged code
26 flagged · loading sourcePackage defines install-time lifecycle scripts.
package.jsonView on unpkgPackage source references child process execution.
openclaw-plugin/index.jsView on unpkg · L2312Findings
Affected versions and remediation
This report applies to grix-connector@4.11.1.
The versions list below shows the verdict for each recorded version.
- Avoid installing grix-connector@4.11.1. Remove it from direct dependencies and check your lockfile for transitive copies.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
- If this version ran, investigate the affected machine and build environment. Rotate credentials it could access and rebuild from a trusted environment.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.