@luckydraw/cumulus@1.0.69
RLM-based CLI chat wrapper for Claude with external history context management
AI Security Review
scanned 15d ago · by lpm-firewall-aiSource inspection found a guarded visitor capability that uses the user’s Codex authentication, which warrants a warning under the agent capability policy. The inspected source does not show a concrete malicious credential exfiltration path or malicious install behavior.
Decision evidence
public snapshot- The gateway can invoke an isolated Codex visitor that copies the user’s Codex auth file into a temporary home.
- The visitor also accepts an explicit test provider, but its base URL is restricted to loopback.
- The package postinstall invokes a package-local GPU-provider cleanup script; this does not establish malicious install behavior.
- The visitor validates a pinned model and allows only the fidas_call tool; it checks that the connected service exposes exactly the configured tool allowlist.
- The visitor’s alternate provider path enforces a loopback URL, and the default path uses the Codex CLI’s existing auth file.
Source & flagged code
9 flagged · loading sourcePackage defines install-time lifecycle scripts.
package.jsonView on unpkgPackage source references child process execution.
dist/tui/hooks/useClaudeProcess.jsView on unpkg · L3Findings
Affected versions and remediation
This report applies to @luckydraw/cumulus@1.0.69.
See all versions of @luckydraw/cumulus for other recorded verdicts.
- Review the evidence and your use of @luckydraw/cumulus@1.0.69 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.