@raolin2025/claude-code-node@3.3.0
Node.js AI Code Agent CLI - Zero dependencies, pure JavaScript, security hardened, multi-channel notifications, Telegram & QQ Bot remote programming, rich media upload, multi-account management
AI Security Review
scanned 1mo ago · by lpm-firewall-aiThe package is a user-started coding-agent CLI with no npm install lifecycle and no self-dependency chain. Shell, publish, scheduler, and notification code is package-aligned and stays behind explicit CLI or tool invocation plus local safety checks.
Decision evidence
public snapshot- package.json exposes only start, dev, test, and repl scripts plus cc-node and cc-notify bins, with no install lifecycle hook.
- src/__tests__/dream.test.js is a node:test unit test for dream summarization and does not run at install.
- The Bash tool calls checkBashSafety and returns a block message before spawning a shell.
- Scheduler exec is a caller-supplied task runner inside createScheduler, not module-load code.
- notify-daemon starts main only when the file is the process entry point.
- NpmPublish queries and publishes only to registry.npmjs.org when its tool action runs.
- ssrf-guard returns blocked for loopback and private IPv4 ranges.
- Enhanced permissions persist rules under the project .claude-code directory.
- Bash, scheduler exec, and NpmPublish can run shell commands or read an npm token only after a user or agent invokes those tools.
- The notify daemon can write a local API key and talk to user-configured Telegram or QQ endpoints when the cc-notify command is started.
Source & flagged code
15 flagged · loading sourcePackage contains a critical-looking secret pattern.
src/__tests__/dream.test.jsView on unpkg · L421GitHub personal access token in src/__tests__/dream.test.js
src/__tests__/dream.test.jsView on unpkg · L421GitHub personal access token in src/__tests__/dream.test.js
src/__tests__/dream.test.jsView on unpkg · L437Findings
Affected versions and remediation
This report applies to @raolin2025/claude-code-node@3.3.0.
See all versions of @raolin2025/claude-code-node for other recorded verdicts.
- Review the evidence and your use of @raolin2025/claude-code-node@3.3.0 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.