pp-numista

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill contains an explicit CLI template that expects an OAuth bearer/token to be supplied as a command-line argument (e.g., numista-pp-cli auth set-token <token>), which would require the agent to accept and embed secret values verbatim, creating an exfiltration risk.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The skill drives the numista-pp-cli which "wraps the Numista REST API" (e.g., catalogues, types, users endpoints) and its Agent Mode explicitly tells agents to parse .results and act on that JSON (and even supports webhook delivery), so the agent will fetch and interpret untrusted public Numista API/user content at runtime.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs installing and running a remote npm package at runtime via "npx -y @mvanhorn/printing-press install numista --cli-only", which will fetch and execute remote code and is a required dependency for the skill.