tako-cli@0.3.10
Tako CLI — unified launcher for AI coding tools (Claude Code, Codex, Gemini) with multi-agent session management
AI Security Review
scanned 4mo ago · by lpm-firewall-aiStatic inspection found high-risk primitives, but they are tied to explicit installer/CLI workflows for a unified AI-coding-tool launcher and no concrete unconsented exfiltration, persistence, destructive action, or install-time payload was confirmed. Scanner credential-exfiltration and runtime-install labels appear noisy because the observed flows are package-aligned provider detection, telemetry/API usage, and self/client updates.
Decision evidence
public snapshot- dist/index.js can read Claude/Codex auth files during subscription provider detection.
- dist/index.js can write Claude statusLine config at runtime and add dangerous launch flags when user selected.
- install.sh/install.ps1 download Bun and install tako-cli on explicit installer execution.
- package.json has no install/postinstall hook; only prepublishOnly build script.
- Network calls are package-aligned: model/quota/announcement APIs, npm registries, Bun downloads, geo mirror selection.
- Credential material read from local AI tools is stored/used as provider authData for launching supported clients; no confirmed exfiltration flow found.
- Runtime package installs/updates target tako-cli or supported AI client packages under ~/.tako.
- Dangerous Claude/Codex flags are gated by remembered user option selection, not silently forced.
Source & flagged code
8 flaggedSource appears to send environment or credential material to an external endpoint.
dist/index.jsView on unpkg · L2Findings
Affected versions and remediation
This report applies to tako-cli@0.3.10.
See all versions of tako-cli for other recorded verdicts.
- Review the evidence and your use of tako-cli@0.3.10 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.