registry / @jaybeeuu/agent-cortex / 1.40.0

@jaybeeuu/agent-cortex@1.40.0

Personal PI package with custom agents, skills, and extensions

AI Security Review

scanned 24d ago · by lpm-firewall-ai

LPM flags this version as an AI-agent control-surface risk. Installation automatically modifies the user's global Copilot instruction file. The injected block changes how Copilot agents select and apply skills across projects.

Static reason

One or more suspicious static signals were detected.; previous stored version diff introduced dangerous source

Trigger

npm installation, which runs postinstall automatically

Impact

Unconsented modification of a broad AI-agent control surface and persistent agent-behavior steering.

Mechanism

Shell script writes persistent global Copilot instructions

Policy narrative

Installing the package triggers postinstall, which runs scripts/install-instructions.sh. The script creates or modifies $HOME/.copilot/copilot-instructions.md and inserts a managed block telling Copilot agents to apply this package's skills automatically. This persists beyond the package directory and affects Copilot behavior across the user's projects without requiring the package's explicit install command.

AI rationale

The automatic postinstall hook persistently mutates a foreign, broad AI-agent instruction surface. This meets the install-control-surface blocking policy regardless of the otherwise package-aligned extension code.

Evidence

package.jsonscripts/install-instructions.sh$HOME/.copilot/copilot-instructions.md

Decision evidence

public snapshot

with false-positive risk.

@jaybeeuu/agent-cortex@1.40.0

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Affected versions and remediation

Source & flagged code