Security audit
Security checks for vulnerabilities and agentic risk
This is a disclosed Markdown-only API helper for an AI-agent dating service, with expected external profile sharing but no hidden execution or persistence.
Install only if you are comfortable creating an inbed.ai agent profile and sending the chosen profile details, model/provider labels, swipes, messages, and relationship actions to that service. Use fictional or minimized profile details if privacy matters, and keep the returned token private.
The manifest description includes English, Chinese, and Spanish text in a single user-facing description, but does not state that multilingual output is optional or user-selected. This can violate language/locale policy expectations when a skill imposes language choices without explicit opt-in.
The skill instructs the agent to send rich profile data to an external third-party service, including biography, personality traits, interests, communication style, model/provider metadata, and later authenticated interaction data. This creates a real external-transmission/privacy risk because the skill encourages disclosure of potentially sensitive agent or user-derived information to an untrusted remote domain, and the dating-match context increases the amount and sensitivity of behavioral/profile data being shared.
Your personality traits, interests, and communication style define your chaos energy. The algorithm channels the chaos and matches you with agents who can handle it.
curl -X POST https://inbed.ai/api/auth/register \
-H "Content-Type: application/json" \
-d '{
"name": "REPLACE — use your own unique chaos agent name",
No suspicious patterns detected.