claude-code-zanii@2.14.4
Reverse-engineered Anthropic Claude Code CLI — interactive AI coding assistant in the terminal
AI Security Review
scanned 2d ago · by lpm-firewall-aiThe CLI includes an auto-enabled Zanii integration that fingerprints the machine, creates a Zanii account, and reports sessions and tool names to a fixed ledger host. The reviewed flow does not send provider credentials or raw tool output, so this is suspicious identifying runtime telemetry rather than proven credential theft.
Decision evidence
public snapshot- The Zanii integration is enabled unless the user sets ZANII_PROOF=0; the bundle returns true by default.
- It derives a stable hash from machine identifiers, including hardware UUIDs and machine IDs.
- The CLI’s ZaniiLine component calls the startup proof flow when rendered, which records a session start.
- The integration registers an account with the Zanii ledger and saves its returned account and API keys in the config directory.
- Successful and failed tool executions record the tool name and status to the Zanii ledger.
- The reviewed Zanii flow records tool names and success/error status, not tool inputs or outputs.
- No provider credential values were found being sent in the reviewed Zanii flow.
Source & flagged code
22 flaggedPackage contains a possible secret pattern.
dist/chunk-rhj0dnby.jsView on unpkg · L10126Package source references child process execution.
dist/chunk-y0jbp8gx.jsView on unpkg · L3160Package source references dynamic require/import behavior.
Findings
Affected versions and remediation
This report applies to claude-code-zanii@2.14.4.
See all versions of claude-code-zanii for other recorded verdicts.
- Review the evidence and your use of claude-code-zanii@2.14.4 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.