carryfeed-twitter-tools

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md explicitly directs the agent to run the CarryFeed CLI to resolve public X/Twitter links and ingest post text, profile bios, and media URLs (public user-generated content) as part of the agent's workflow and decision rules, so the agent will read untrusted third‑party social content that could contain instructions influencing subsequent actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly requires running the remote CLI via "npx -y @carryfeed/cli", which at runtime fetches and executes code from the npm package (https://www.npmjs.com/package/@carryfeed/cli), so the skill depends on and runs externally fetched code.