getform

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

• Third-party content exposure detected (high risk: 0.80). The skill’s SKILL.md instructs the agent to connect to Getform via the Membrane CLI and to run/discover actions (e.g., membrane action run / action list) that return form submissions and other Getform data—which are user-generated, public/untrusted third-party content the agent is expected to read and act on.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill relies on fetching and running the Membrane CLI at runtime (e.g., "npm install -g @membranehq/cli@latest" and "npx @membranehq/cli@latest"), which downloads and executes remote code that the skill uses as a required dependency.