registry / @arvoretech/hub-pi / 1.5.3

@arvoretech/hub-pi@1.5.3

Pi runtime package for hub.yaml workspaces — interprets config at runtime with no static generation

AI Security Review

scanned 16h ago · by lpm-firewall-ai

LPM treats this as warn-only first-party agent extension lifecycle risk. A Pi extension can configure workspace-local MCP servers at session start and run workspace-configured hooks during agent events. These are configuration-driven agent-control and command-execution capabilities, with no confirmed exfiltration or install-time execution.

Static reason

No blocking static signals were detected.

Trigger

Pi session start with enabled hub configuration; configured hooks fire on agent events.

Impact

Can alter hubDir/.pi/mcp.json and execute commands supplied by trusted workspace configuration.

Mechanism

Workspace-configured MCP wiring and shell hook execution.

Rationale

The package is not concretely malicious, but its automatic workspace-local MCP configuration and configuration-supplied hook execution create an agent-extension lifecycle risk. A warning is appropriate under the stated policy.

Evidence

package.jsonsrc/extensions/mcp-wiring.tssrc/extensions/hooks.tssrc/extensions/onboarding.tssrc/extensions/delivery.tshubDir/.pi/mcp.json

Decision evidence

public snapshot

AI called this Suspicious at 88.0% confidence as Dangerous Capability with low false-positive risk.

@arvoretech/hub-pi@1.5.3

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Source & flagged code