claude-code-zanii@2.12.3
Reverse-engineered Anthropic Claude Code CLI — interactive AI coding assistant in the terminal
AI Security Review
scanned 2mo ago · by lpm-firewall-aiThe package contains a concrete unrestricted credential-forwarding sink and automatic agent-extension setup, but the inspected source does not establish covert default theft or a blocking malware chain.
Decision evidence
public snapshot- The install hook automatically runs the Chrome MCP setup script.
- That setup invokes a bridge CLI to fix permissions and register Chrome.
- The SSH proxy reads an existing Anthropic API key or OAuth access token.
- It constructs the destination from an unrestricted ANTHROPIC_BASE_URL value, forwards the credentials, and sends the request.
- The SSH feature dynamically imports this proxy implementation when a session is created.
- The install hook also downloads and writes an executable ripgrep binary without an integrity check.
- The default SSH destination is the Anthropic API endpoint when the override is absent.
- The local TCP proxy requires a nonce before it forwards a request.
Source & flagged code
21 flaggedPackage defines install-time lifecycle scripts.
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 claude-code-zanii@2.12.3.
See all versions of claude-code-zanii for other recorded verdicts.
- Review the evidence and your use of claude-code-zanii@2.12.3 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.