@eventmodelers/cli@2.0.27
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 5d ago · by lpm-firewall-aiThe package exposes a CLI without install hooks. Its credential use targets the configured Eventmodelers API, while the unofficial agent endpoint is optional and user-selected; project agent configuration changes occur during explicit setup.
Decision evidence
public snapshot- The manifest has no install lifecycle hooks; it exposes cli.js as a command-line executable.
- The default board API is api.eventmodelers.ai, and the CLI sends the Eventmodelers token to the configured board API for its intended account operation.
- The Franconia endpoint is an optional Mastra endpoint; its API key comes from MASTRA_API_KEY, and the endpoint is set from the user's --endpoint argument.
- Prototype setup is an explicit init operation that installs project skills and permission settings; the source says Claude is started by the user.
- The CLI can execute shell commands and run agent harnesses, but these are user-selected runtime features of the agent CLI.
- The prototype extension writes Claude project configuration, but only as part of its explicit setup workflow.
Source & flagged code
12 flaggedPackage source references child process execution.
stacks/supabase/templates/root/src/common/testHelpers.tsView on unpkg · L1This package version adds a dangerous source file absent from the previous stored version; route for source-aware review.
shared/build-kit/ralph-exec.jsView on unpkgFindings
Affected versions and remediation
This report applies to @eventmodelers/cli@2.0.27.
See all versions of @eventmodelers/cli for other recorded verdicts.
- Review the evidence and your use of @eventmodelers/cli@2.0.27 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.