@zaip-ai/cli@1.3.2
CLI oficial para instalar e configurar o ZAIP-AIOS em qualquer repositório
AI Security Review
scanned 2mo ago · by lpm-firewall-aiThis is an explicit setup tool with broad AI-agent configuration effects, not a demonstrated malware chain. The configuration capability warrants a warning rather than a publication block.
Static reason
No blocking static signals were detected.
Trigger
An operator runs zaip init without --project-only.
Impact
Future Codex sessions in the initialized project can use unrestricted filesystem writes and full network access under the generated profile.
Mechanism
Writes agent configuration, hooks, and a broad Codex permission profile.
AI rationale
This is an explicit setup tool with broad AI-agent configuration effects, not a demonstrated malware chain. The configuration capability warrants a warning rather than a publication block.
Evidence
dist/commands/init.jsdist/services/global-codex-config.jsdist/services/adapters.jsdist/services/codex-config-template.js.agents/**.codex/config.toml.codex/hooks/zaip-pre-tool-use.mjs.codex/hooks/zaip-subagent-lifecycle.mjs~/.codex/config.toml
Network endpoints1
developers.openai.com/mcp
Decision evidence
public snapshotAI called this Suspicious at 92.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- The user-invoked init command enables a global Codex configuration change unless project-only mode is selected.
- The installer reads and replaces the user's global Codex config to add a command hook.
- The generated project Codex configuration defaults to unrestricted filesystem writes and full network access.
Evidence against
- There are no npm preinstall, install, or postinstall hooks.
- The configuration changes occur only after an explicit CLI init command.
- The shipped Windows executable has corresponding source that launches the local design runtime; no secret collection or exfiltration was found.
- The document and executable findings are template assets, not loaded by the doctor command.
Behavioral surface
ChildProcessCryptoEnvironmentVarsFilesystemShell
HighEntropyStringsUrlStrings
Source & flagged code
10 flaggedtemplates/.agents/design/zaip-design.exeView file
•path = templates/.agents/design/zaip-design.exe
kind = payload_in_excluded_dir
sizeBytes = 7680
magicHex = [redacted]
High
Payload In Excluded Dir
Package hides binary, compressed, or executable-looking payloads in test/fixture/hidden paths.
templates/.agents/design/zaip-design.exeView on unpkg•path = templates/.agents/design/zaip-design.exe
kind = native_binary
sizeBytes = 7680
magicHex = [redacted]
Medium
Ships Native Binary
Package ships native binary artifacts.
templates/.agents/design/zaip-design.exeView on unpkgtemplates/.agents/design/packages/windows-launcher/build-design-launcher.ps1View file
•path = templates/.agents/design/packages/windows-launcher/build-design-launcher.ps1
kind = build_helper
sizeBytes = 3452
magicHex = [redacted]
Findings
1 Critical3 High7 Medium6 Low
CriticalTrigger Reachable Excluded Directory Payloaddist/services/doctor.js
HighShips High Entropy Blobtemplates/.agents/branding/brandbook.docx
HighPayload In Excluded Dirtemplates/.agents/design/zaip-design.exe
HighSemantic Analysis Limiteddist/services/codex-pre-tool-hook-template.js#virtual:normalized:round1
MediumEnvironment Vars
MediumShips Native Binarytemplates/.agents/design/zaip-design.exe
MediumShips Build Helpertemplates/.agents/design/packages/windows-launcher/build-design-launcher.ps1
MediumShips Compressed Blobtemplates/.agents/branding/brandbook.docx
MediumOversized Source Filetemplates/.agents/runtime/bin/zaip.mjs
MediumOversized Cli Entrypointtemplates/.agents/runtime/bin/zaip.mjs
Affected versions and remediation
This report applies to @zaip-ai/cli@1.3.2.
See all versions of @zaip-ai/cli for other recorded verdicts.
- Review the evidence and your use of @zaip-ai/cli@1.3.2 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.