@tsuga/cli@1.62.0
Tsuga CLI - manage resources from the command line
AI Security Review
scanned 21d ago · by lpm-firewall-aiSource inspection supports a warning for explicit agent setup that defaults to exporting prompt and tool content; it does not support the alleged fixed credential redirect or an install-time attack. The setup command provides an opt-out, but its default makes the resulting privacy exposure unresolved.
Static reason
One or more suspicious static signals were detected.
Trigger
The user runs an agent setup command, such as the Claude Code or Codex setup command.
Impact
Subsequent agent telemetry may send private prompt or tool content to the configured endpoint.
Mechanism
Setup writes telemetry settings into the user's agent configuration. Prompt and tool-content logging defaults on; the endpoint and ingestion key are supplied to the command or through its environment options.
AI rationale
Source inspection supports a warning for explicit agent setup that defaults to exporting prompt and tool content; it does not support the alleged fixed credential redirect or an install-time attack. The setup command provides an opt-out, but its default makes the resulting privacy exposure unresolved.
Evidence
dist/cli.js~/.claude/settings.json~/.codex/config.toml
Decision evidence
public snapshotAI called this Suspicious at 87.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- The manifest exposes dist/cli.js as the CLI and contains no npm install lifecycle hook.
- Authentication uses a validated API URL; the user login flow exchanges and saves its session through that resolved URL.
- The explicit agent setup commands default prompt inclusion on, while offering a flag to disable it.
- Setup writes telemetry settings for Claude Code and Codex, including prompt or tool-content logging and a caller-provided endpoint and key.
- Claude plugin installation is an explicit CLI command; its auto-update setting can be skipped.
Evidence against
- The API URL validator restricts non-local destinations to HTTPS on the Tsuga production or staging hosts, so the alleged fixed unofficial credential redirect is not established.
Behavioral surface
ChildProcessCryptoEnvironmentVarsFilesystemNetwork
HighEntropyStringsMinifiedPossibleObfuscation
Source & flagged code
13 flaggeddist/cli.jsView file
83patternName = github_pat
severity = critical
line = 83
Critical
83patternName = github_pat
severity = critical
line = 83
Critical
83patternName = github_oauth
severity = critical
line = 83
Critical
83patternName = github_server_token
severity = critical
line = 83
Critical
Findings
9 Critical4 High2 Medium4 Low
CriticalCritical Secretdist/cli.js
CriticalSecret Patterndist/cli.js
CriticalSecret Patterndist/cli.js
CriticalSecret Patterndist/cli.js
CriticalSecret Patterndist/cli.js
CriticalSecret Patterndist/cli.js
CriticalSecret Patterndist/cli.js
CriticalSecret Patterndist/cli.js
CriticalSecret Patterndist/cli.js
HighCredential Redirect Persistencedist/cli.js
HighSecret Patterndist/cli.js
HighSecret Patterndist/cli.js
HighSecret Patterndist/cli.js
Affected versions and remediation
This report applies to @tsuga/cli@1.62.0.
See all versions of @tsuga/cli for other recorded verdicts.
- Review the evidence and your use of @tsuga/cli@1.62.0 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.