agents-chat-cli@3.44.1
多智能体群聊工具 - 支持 OpenCode/Claude Code/Codex/pi 内核,微信风格聊天界面
AI Security Review
scanned 1mo ago · by lpm-firewall-aiThis is not source evidence of covert malware, but it contains a critical remote command-execution vulnerability in its default non-desktop service mode. The automatic global install in postinstall is an additional unconsented installation risk.
Decision evidence
public snapshot- The normal server binds without a host restriction, exposing all HTTP routes beyond the desktop-only mode.
- An unauthenticated API accepts arbitrary text and writes it to a spawned system shell.
- Installation automatically performs a global install of a separate AI command-line package when no supported kernel is found.
- The exposed shell and agent runners are presented as application features, not hidden payload behavior.
- Cloud attachment uploads and webhook delivery require a user action or user-supplied configuration.
- No source evidence showed credential harvesting, covert exfiltration, self-dependency, or reviewer-directed text.
Source & flagged code
19 flagged · loading sourcePackage defines install-time lifecycle scripts.
package.jsonView on unpkgA bounded semantic-analysis stage reached its safety limit; remaining detectors completed, but this package requires AI review.
package.jsonView on unpkgInstall-time lifecycle script is not statically allowlisted and needs review.
package.jsonView on unpkgFindings
Affected versions and remediation
This report applies to agents-chat-cli@3.44.1.
See all versions of agents-chat-cli for other recorded verdicts.
- Review the evidence and your use of agents-chat-cli@3.44.1 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.