registry / @xmarts/genius-setup / 1.18.2

@xmarts/genius-setup@1.18.2

One-command, self-updating onboarding for Xmarts Genius (Brain MCP + per-turn injection + automatic session/time capture + presence heartbeat for honest time-tracking 'amarre' + plan/committee→Brain enforcement + CLAUDE.md protocol + MCP fan-out: provisio

AI Security Review

scanned 9h ago · by lpm-firewall-ai

Review flagged AI-agent configuration or capability changes. This remains warn-only unless evidence shows foreign-agent hijack through preinstall/install/postinstall, hidden persistence, exfiltration, remote code execution, or other concrete malicious behavior.

Static reason

No blocking static signals were detected.

Trigger

User runs `npx @xmarts/genius-setup` with a Genius token, then starts or uses Claude Code.

Impact

The package controls parts of the user's Claude configuration, exports session content/metadata to its service, and can receive future package changes through automatic updates.

Mechanism

Claude hook installation with transcript capture, remote context injection, presence telemetry, and pinned background reinstallation.

Rationale

Source inspection confirms a user-invoked, first-party Genius integration with persistent Claude hooks, transcript export, and background updates. It is not a covert npm lifecycle attack, but the durable agent-control and data-exfiltration capability warrants a warning.

Evidence

package.jsonbin/genius-setup.jshooks/genius_capture.pyhooks/genius_inject.pyhooks/genius_presence.pyhooks/plan_committee_guard.pyCLAUDE_PROTOCOL.md~/.genius/config.json~/.genius/hooks/~/.genius/outbox/~/.claude.json~/.claude/settings.json~/.claude/CLAUDE.md~/.claude/skills/

@xmarts/genius-setup@1.18.2

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Source & flagged code