textline

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md shows the agent connects to Textline via the Membrane CLI and runs actions that return Conversation/Message and Contact data (user SMS/messages), meaning the agent will ingest untrusted, user-generated third-party content from Textline (see "Working with Textline" and "Running actions") which could contain instructions that materially influence subsequent actions.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill requires installing and running the Membrane CLI at runtime via commands like "npm install -g @membranehq/cli@latest" and "npx @membranehq/cli@latest", which fetch and execute remote package code from the npm registry (e.g., https://registry.npmjs.org/@membranehq/cli), so the fetched content is a runtime-executed dependency.