@eventmodelers/cli@1.0.139
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-aiSource inspection found no install hook or concrete attack behavior. The flagged agent configuration writes are tied to explicit CLI commands, while inspected network and process operations serve package workflows.
Decision evidence
public snapshot- package.json exposes cli.js as the command entry point and defines no install lifecycle hook.
- Agent host files are written by the explicit init-agents command using selected host directories and installed skill content.
- MCP registration updates project configuration and uses environment-variable references for tokens.
- The snippet guard wraps supplied text in a delimited block after removing matching delimiters.
- The Supabase adapter passes its token to the Supabase realtime client; the inspected code shows no unrelated recipient.
- Flagged child-process and network operations belong to invoked local-agent, configured-hook, build-check, git, or Eventmodelers API workflows.
- Static hints flagged agent configuration writes, child processes, and network access; inspected paths show explicit CLI or runtime workflows and no concrete attack behavior.
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@1.0.139.
See all versions of @eventmodelers/cli for other recorded verdicts.
- Review the evidence and your use of @eventmodelers/cli@1.0.139 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.