@luckydraw/cumulus@1.0.30
RLM-based CLI chat wrapper for Claude with external history context management
AI Security Review
scanned 3d ago · by lpm-firewall-aiThe scanner flags point to normal Cumulus features: Claude subprocess use, message routing, optional service setup, and explicit package updates. The inspected source did not establish malicious behavior or an unresolved attack surface.
Decision evidence
public snapshot- package.json runs Husky in its prepare hook and lists dependencies without a runtime dependency on itself.
- Content analysis runs the Claude CLI with a prompt sent through standard input and filters Claude-related environment variables.
- The inspected runtime operations format thread messages, collect context, and manage tool and worktree state.
- Gateway callers pass messages from users, configured chat adapters, webhooks, scheduled triggers, and queued jobs into thread processing.
- The gateway agent MCP entry is omitted unless gateway configuration is supplied; its API URL and key come from that configuration.
- Service installation asks for interactive confirmation, and the updater installs a selected version of this package when invoked.
- The package uses child processes, webhooks, chat adapters, scheduled agent triggers, and an npm updater; inspected code ties these to its gateway and CLI functions.
Source & flagged code
8 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@1.0.30.
See all versions of @luckydraw/cumulus for other recorded verdicts.
- Review the evidence and your use of @luckydraw/cumulus@1.0.30 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.