@zaip-ai/cli@3.1.0
CLI oficial para instalar e configurar o ZAIP-AIOS em qualquer repositório
AI Security Review
scanned 2mo ago · by lpm-firewall-aiThis is a user-invoked agent and infrastructure management tool with a broad permission profile and opaque binary payload. No lifecycle execution, self-dependency chain, automatic payload activation, credential exfiltration, or malicious install hook was found.
Decision evidence
public snapshot- The explicit init command enables a user-level Codex configuration update unless project-only mode is selected.
- The configuration reconciler writes a PreToolUse hook into the Codex home configuration.
- The installed project configuration defaults to unrestricted filesystem and network access.
- The package ships an opaque Windows executable in its project template.
- The cloud runner reads project credentials and can perform contract-gated DNS, Terraform, and SSH operations.
- package.json has no preinstall, install, or postinstall lifecycle hook and no self-dependency.
- The global configuration change is reached from the user-invoked init command, not package installation or import.
- The cloud runner requires an authorization-bearing contract, exact target checks, and credential inputs.
- The doctor source only checks required project files; it does not load or execute the flagged design executable.
Source & flagged code
8 flaggedSource gates dangerous network, credential, or execution behavior behind CI, host, platform, time, or geo fingerprint checks.
dist/services/stack-git.jsView on unpkg · L1Package hides binary, compressed, or executable-looking payloads in test/fixture/hidden paths.
templates/.agents/design/zaip-design.exeView on unpkgFindings
Affected versions and remediation
This report applies to @zaip-ai/cli@3.1.0.
See all versions of @zaip-ai/cli for other recorded verdicts.
- Review the evidence and your use of @zaip-ai/cli@3.1.0 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.