registry / adghost-lily / 3.8.0

adghost-lily@3.8.0

Lily (built by AdGhost) - an enforced, multi-IDE AI engineering skill pack: status-line output contract, executable security / UI-slop / minimal-diff gates, persistent memory, and a 20-stage router. Installs into Claude Code, Cursor, Codex, Antigravity, C

AI Security Review

scanned 19h 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 the package bin or bundled installer script.

Impact

Agent behavior and project-local hook configuration can change; hooks write local memory/journal files and run local checks.

Mechanism

Copies agent configuration and optionally installs local journal/security hooks.

Rationale

This is a user-invoked agent-extension installer rather than malicious npm lifecycle behavior. It warrants a warning because it mutates AI-agent control surfaces and can establish hooks in the target project.

Evidence

package.jsonbin/install.mjs.claude/settings.jsonscripts/install.shscripts/setup-hooks.pyscripts/lily-journal.py.cursor/hooks.json.agents/hooks.json.git/hooks/pre-commit.memory/session-journal.md

Decision evidence

public snapshot

AI called this Suspicious at 91.0% confidence with false-positive risk.

adghost-lily@3.8.0

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Source & flagged code