tako-cli@0.3.9
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-aiSource inspection did not confirm credential exfiltration, install-time malware, or destructive behavior; most network and execution primitives are aligned with an AI coding-tool launcher. The unprompted Claude Code settings write and auth-material import keep this above clean.
Decision evidence
public snapshot- dist/index.js writes Claude Code settings statusLine to run the Tako statusline command on normal interactive launch
- dist/index.js reads local Claude and Codex auth state and stores provider authData in Tako config for launcher use
- dist/index.js enables selected dangerous flags for Claude/Codex when saved options include skip-permissions or bypass-sandbox
- dist/index.js sends telemetry/identify events, including email when available, to https://op.shiroha.tech/api
- package.json has no install/postinstall hook; prepublishOnly is publisher-side build only
- install.sh/install.ps1 are explicit user-run installers for Bun and tako-cli, not npm lifecycle execution
- Runtime package installs are user-invoked update/install flows under ~/.tako/tools or ~/.tako/cli
- Provider Authorization headers are used for configured model/quota/announcement APIs, not broad env dumping
- No evidence of destructive behavior, persistence beyond user PATH/wrapper setup, or arbitrary remote code payloads
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@0.3.9.
See all versions of tako-cli for other recorded verdicts.
- Review the evidence and your use of tako-cli@0.3.9 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.