registry / @mesosfer-org/ozon-cli / 1.4.7

@mesosfer-org/ozon-cli@1.4.7

CLI Agent for Ozon

AI Security Review

scanned 5h ago · by lpm-firewall-ai

Review flagged AI-agent configuration or capability changes. This remains warn-only unless evidence shows foreign-agent hijack through preinstall/install/postinstall, hidden persistence, exfiltration, remote code execution, or other concrete malicious behavior.

Static reason

No blocking static signals were detected.

Trigger

User runs ozon-agent chat, run, bg, or install-skill.

Impact

A compromised backend, malicious skill, or prompt-injected workspace could cause project changes or shell commands in the user-selected workspace.

Mechanism

Remote LLM/marketplace content drives project-scoped tools, shell commands, and skill writes.

Rationale

Source inspection found no concrete malicious behavior or lifecycle execution, but did find an autonomous LLM-to-shell/file-write capability and remote skill installation. Flag as a warning for dangerous dual-use agent capability rather than block.

Evidence

package.jsondist/index.jsdist/commands/bg.jsdist/commands/chat.jsdist/utils/background.jsdist/utils/skills.jsdist/utils/tools.js.ozon/sessions/<id>/task.json.ozon/sessions/<id>/status.json.ozon/skills/<name>.md~/.ozon/skills/<name>.md

Network endpoints4

api.ozon.my.id

@mesosfer-org/ozon-cli@1.4.7

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Source & flagged code