@vatio-ai/skills@1.29.2
Vatio skills for Claude Code, Cursor and Codex: add an AI agent to the app you're building with /vatio-init, then improve it daily with /vatio-improve and learn what people ask it with /vatio-analyze.
AI Security Review
scanned 6d ago · by lpm-firewall-aiThe agent-configuration writes are the package’s stated purpose and occur only when its CLI is explicitly run, within the selected project. Its disclosed installation telemetry is opt-out and sends package version, coding-agent type, and a session ID to Vatio’s own endpoint; no malicious behavior was identified.
Decision evidence
public snapshot- The manifest exposes bin/install.mjs as the vatio-skills command; it has no lifecycle scripts.
- When invoked, the installer copies the package’s skills into .agents/skills and .claude/skills under the selected project directory.
- The installer discloses an installation event and supports opt-out through VATIO_TELEMETRY=0 or DO_NOT_TRACK.
- The event goes to Vatio’s /api/developer/v1/cli/events endpoint and reports the skills version, coding-agent type, and a session ID.
- The README describes these skills as instructions with no hooks, MCP servers, or automatic behavior.
- The session ID is stored locally in ~/.vatio/sessions.json and is generated or reused for the project session.
- The installation event is part of the explicitly invoked CLI command, not an install lifecycle hook.
Source & flagged code
8 flaggedRuntime or CLI source writes behavior-bearing configuration into a user or project AI-agent control surface.
bin/install.mjsView on unpkgFindings
Affected versions and remediation
This report applies to @vatio-ai/skills@1.29.2.
See all versions of @vatio-ai/skills for other recorded verdicts.
- Review the evidence and your use of @vatio-ai/skills@1.29.2 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.