@luckydraw/cumulus@0.31.61
RLM-based CLI chat wrapper for Claude with external history context management
AI Security Review
scanned 3mo ago · by lpm-firewall-aiNo automatic malicious lifecycle behavior was confirmed, so blocking is not justified. The forced permission bypass combined with local-file and outbound agent tools warrants a warning.
Decision evidence
public snapshot- dist/lib/gateway.js spawns Claude with bypassPermissions.
- dist/tui/hooks/useClaudeProcess.js also forces bypassPermissions.
- dist/lib/gateway.js injects mandatory agent instructions and creates MCP configs.
- dist/gateway/gateway-agents-mcp.js lets the agent read arbitrary local files for upload.
- MCP can send data to api.resend.com and plastic.luckydraw.dev.
- Agent scheduling enables recurring system-level prompt injection.
- package.json has no preinstall, install, or postinstall hook.
- Gateway setup and service persistence require explicit cumulus-gateway setup plus an interactive confirmation.
- Observed network actions are runtime features, not import/install-time behavior.
- No confirmed credential harvesting or covert exfiltration path was found.
Source & flagged code
7 flagged · loading sourcePackage source references child process execution.
dist/tui/hooks/useClaudeProcess.jsView on unpkg · L3This package version adds a dangerous source file absent from the previous stored version; route for source-aware review.
dist/lib/gateway.jsView on unpkgFindings
Affected versions and remediation
This report applies to @luckydraw/cumulus@0.31.61.
See all versions of @luckydraw/cumulus for other recorded verdicts.
- Review the evidence and your use of @luckydraw/cumulus@0.31.61 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.