@eventmodelers/cli@1.0.155
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-aiThe manifest has no install hook, and source inspection found the flagged runtime and configuration behaviors tied to the CLI’s project workflows and package service. No concrete attack surface was established.
Decision evidence
public snapshot- package.json declares a CLI entrypoint and no install lifecycle hooks.
- The agent-host mapping writes skill stubs into project directories; MCP registration uses project configuration and an environment-variable token reference.
- The runtime realtime adapter subscribes to caller-supplied topics and handles connection status and retries; the CLI wires it to board and prompt events for the configured agent.
- The local AI launcher starts its companion process in the chosen project directory; the bridge hook runs only when configured, and the typecheck helper runs a project build check.
- The code exporter scopes its files to configured slice and workspace paths; the Spec Kitty adapter uses the package service endpoint.
- The package-owned MCP configuration uses the configured service base URL and token reference, and its enablement writes project-local settings.
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@1.0.155.
See all versions of @eventmodelers/cli for other recorded verdicts.
- Review the evidence and your use of @eventmodelers/cli@1.0.155 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.