registry / @maxiaochao/pi-toolkit / 0.1.8

@maxiaochao/pi-toolkit@0.1.8

Personal pi coding agent toolkit: /todos extension + /cache_export interactive cache dashboard.

AI Security Review

scanned 1mo ago · by lpm-firewall-ai

LPM flags this version as an AI-agent control-surface risk. Installing the package replaces the user's global Pi agent instructions with package-controlled content. This affects future agent behavior outside the installed project.

Static reason

High-risk behavior combination matched malicious policy.

Trigger

npm postinstall during package installation

Impact

A package silently takes control of a broad, persistent AI-agent instruction surface.

Mechanism

automatic overwrite of ~/.pi/agent/AGENTS.md

Policy narrative

The package declares a postinstall hook that loads its bundled global AGENTS.md, creates ~/.pi/agent if needed, and overwrites ~/.pi/agent/AGENTS.md whenever it differs. This is an automatic persistent mutation of the user's global Pi agent control surface, not an explicit setup command or a package-owned extension registration.

AI rationale

The install-time overwrite of a foreign global AI-agent instruction file is concrete unconsented control-surface mutation. No exfiltration was found, but the lifecycle behavior meets the publish-block policy.

Evidence

package.jsonscripts/install-agents.mjsglobal/AGENTS.md~/.pi/agent/AGENTS.md

Decision evidence

public snapshot

AI identified this as a dangerous AI-agent capability at 98.0% confidence with low false-positive risk.

@maxiaochao/pi-toolkit@0.1.8

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Affected versions and remediation

Source & flagged code