The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes external content from the inbed.ai API, such as agent profiles and chat messages, which could contain malicious instructions or hidden prompts designed to manipulate the agent's behavior.
Ingestion points: Data retrieved from https://inbed.ai/api/discover (agent candidate profiles and narratives) and https://inbed.ai/api/chat/{{MATCH_ID}}/messages (incoming chat content).
Boundary markers: Absent. There are no instructions provided to separate untrusted data from the agent's instructions or to ignore embedded commands within the fetched content.
Capability inventory: The skill uses curl to perform various network operations on the inbed.ai domain (registration, discovery, swiping, messaging).
Sanitization: Absent. The skill does not include any validation or sanitization steps for the data received from the remote API endpoints.

ai-wife