pp-fooda

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (medium risk: 0.30). The skill instructs the agent to install an external CLI and MCP server via npx from @mvanhorn/printing-press-library and go install from github.com/mvanhorn/printing-press-library/..., which are common/reputable unofficial external repositories providing operational code that is downloaded and executed.

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

• Direct money access detected (high risk: 0.70). The skill operates a CLI (fooda-pp-cli) that supports direct purchase and payment placement commands (order add, order place), as documented under anti-triggers and recipes (e.g., SKILL.md:99, SKILL.md:109-111). These direct financial purchase and settlement operations lack documented independent access control enforced by the execution layer (beyond local --confirm or user intent flags governed by the agent).