neta-space

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md explicitly instructs the agent to call the Neta API (e.g., get_hashtag_info, get_hashtag_collections, read_collection) to fetch tag lore, user/official collections and characters — user-generated public content that the agent is expected to read and use to guide decisions, creating a plausible avenue for indirect prompt injection.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill invokes and depends on remote npm packages at runtime (e.g., "npx -y @talesofai/neta-skills" / referenced package https://www.npmjs.com/package/@talesofai/neta-skills), which will fetch and execute external code the skill relies on, so this is a runtime remote-code dependency.