registry / izanagi-ai / 2.0.4

izanagi-ai@2.0.4

Modular Skill-Oriented AI Prompt & Agent Framework for Autonomous Software Engineering

AI Security Review

scanned 4h ago · by lpm-firewall-ai

LPM blocks this version under the AI-agent control-surface policy. npm postinstall mutates the consuming project's generic .agents AI-agent control surface. It drops framework instructions and agent assets without an explicit user command.

Static reason

One or more suspicious static signals were detected.

Trigger

npm installation of izanagi-ai@2.0.4

Impact

Can override or introduce AI-agent instructions and configuration in the host project.

Mechanism

postinstall recursive copy into INIT_CWD/.agents

Policy narrative

The install lifecycle loads the package installer, which targets the consumer project via INIT_CWD or the current directory. It recursively copies a large agent framework into the generic .agents directory and overwrites same-named instruction files, causing unconsented modification of a broad AI-agent control surface.

Rationale

This is a concrete unconsented postinstall mutation of a generic host AI-agent directory, not merely an explicit CLI setup command. No additional exfiltration behavior was found, but the lifecycle control-surface write meets the block boundary.

Evidence

package.jsonbin/postinstall.cjsdist/postinstall.jsdist/installer.js.agents/AGENTS.md.agents/SYSTEM.md.agents/RULES.md.agents/agents.agents/skills

izanagi-ai@2.0.4

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Source & flagged code