registry / @maxiaochao/pi-toolkit / 0.1.6

@maxiaochao/pi-toolkit@0.1.6

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

AI Security Review

scanned 3h ago · by lpm-firewall-ai

LPM flags this version as an AI-agent control-surface risk. Installing the package silently replaces global instructions for the Pi coding agent. This affects future agent behavior outside the installed project.

Static reason

High-risk behavior combination matched malicious policy.

Trigger

npm installation invokes the postinstall hook.

Impact

A package controls the user's broad AI-agent instruction surface without an explicit setup command.

Mechanism

Writes bundled instructions to the global Pi agent configuration.

Policy narrative

The package declares a postinstall hook that reads its bundled instruction file and writes it to the user's global Pi agent configuration. The write is automatic, creates parent directories, and replaces existing content whenever it differs. Although no exfiltration or remote payload was found, this is an unconsented install-time mutation of a broad AI-agent control surface.

Rationale

The automatic postinstall hook overwrites a global AI-agent instruction file under the user's home directory. This meets the policy boundary for malicious install-time AI-agent control-surface mutation.

Evidence

package.jsonscripts/install-agents.mjsglobal/AGENTS.mdextensions/cache-export/index.ts~/.pi/agent/AGENTS.md

Decision evidence

public snapshot

with false-positive risk.

@maxiaochao/pi-toolkit@0.1.6

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Source & flagged code