@darkrei08/wizard-ai-cli
Installer for the Wizard-AI environment: AI CLI tools and Claude Code skills (graphify, llmlingua, flashrank, markitdown and more). Clones the repo and runs the platform setup script.
This report shows the evidence for @darkrei08/wizard-ai-cli@0.52.50. 1 of 19 known versions is blocked or flagged. See all versions.
AI Security Review
scanned 3mo ago · by lpm-firewall-aiSource confirms a concrete, automatic chain from the package’s default installer flow to foreign agent-config rewrites, credential/proxy provisioning, and persistence. The lack of an npm lifecycle hook does not mitigate the unsafe default CLI behavior.
Decision evidence
public snapshot- scripts/wizard-installer.js auto-runs proxy auto-setup after installs.
- scripts/wz-ai-proxy.js overwrites ~/.pi/agent auth.json and models.json.
- Proxy redirects Pi providers to local rotator endpoints and provisions OAuth accounts.
- Proxy installs boot-persistent systemd, LaunchAgent, or Windows Startup services.
- scripts/inject-global-prompts.sh writes global Pi, Claude, and Amp instructions.
- Default non-TTY installer installs all registry repos, then performs proxy setup.
- package.json has no npm lifecycle hook.
- Network cloning and setup are reached through CLI execution, not package installation.
Source & flagged code
5 flagged · loading sourceSource writes installer persistence such as shell profile or service configuration.
scripts/wz-ai-proxy.jsView on unpkg · L2Findings
Affected versions and remediation
This report applies to @darkrei08/wizard-ai-cli@0.52.50.
The versions list below shows the verdict for each recorded version.
- Avoid installing @darkrei08/wizard-ai-cli@0.52.50. 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.