@luckydraw/cumulus@1.0.0
RLM-based CLI chat wrapper for Claude with external history context management
AI Security Review
scanned 3mo ago · by lpm-firewall-aiThe package is a deliberately powerful local agent gateway with unrestricted execution primitives. Its persistence and update paths are explicit user-command setup/update flows, not automatic installation behavior.
Decision evidence
public snapshot- dist/lib/gateway.js starts Claude with --permission-mode bypassPermissions.
- dist/lib/tools/bash.js executes arbitrary bash commands.
- dist/lib/tools/write.js can create or overwrite arbitrary paths.
- dist/lib/tools/web-fetch.js permits arbitrary HTTP methods and URLs.
- dist/gateway/server.js exposes these capabilities through an authenticated gateway.
- package.json has no preinstall, install, or postinstall hook.
- dist/gateway/setup.js installs persistence only after explicit setup and TTY confirmation.
- dist/gateway/auth.js requires a configured API key for non-health API routes.
- No hidden credential harvesting or unsolicited exfiltration was found in inspected paths.
Source & flagged code
7 flagged · loading sourcePackage source references child process execution.
dist/tui/hooks/useClaudeProcess.jsView on unpkg · L3Findings
Affected versions and remediation
This report applies to @luckydraw/cumulus@1.0.0.
See all versions of @luckydraw/cumulus for other recorded verdicts.
- Review the evidence and your use of @luckydraw/cumulus@1.0.0 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.