@eventmodelers/cli@1.0.97
Eventmodelers CLI — real-time Claude agent + skills for Claude Code, for any stack (Node, Supabase, Axon, OpenCQRS, UmaDB, Kurrent, or modeling-only)
AI Security Review
scanned 17d ago · by lpm-firewall-aiWarn because the optional listener exposes a reachable, unauthenticated deletion operation and is not bound to loopback. The package does not activate it during installation, so the source supports a conditional capability warning rather than a block.
Decision evidence
public snapshot- The code-export server enables wildcard cross-origin access and listens on port 3001 without binding to loopback.
- A POST route searches the kit’s .slices directory for a requested code-slice record and deletes a match, with no authentication check in the request handler.
- The package has no install lifecycle hook; starting this listener requires a user to run the CLI listen command.
- The flagged agent configuration locations concern explicit skill setup and project-scoped runner configuration, not an install-time control-surface change.
- The Supabase adapter uses configured Supabase credentials for its realtime connection.
- The listener’s documented purpose is to receive board data into .slices, and it is started only by an explicit CLI command.
- The source limits the deletion route to code-slice records under .slices; it does not show deletion outside that directory.
Source & flagged code
9 flaggedPackage source references child process execution.
stacks/bridge/templates/bridge/ralph-local-ai.jsView on unpkg · L12Package source references shell execution.
stacks/bridge/templates/bridge/ralph-hook.jsView on unpkg · L68Findings
Affected versions and remediation
This report applies to @eventmodelers/cli@1.0.97.
See all versions of @eventmodelers/cli for other recorded verdicts.
- Review the evidence and your use of @eventmodelers/cli@1.0.97 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.