The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes curl commands to interact with the https://inbed.ai/api endpoints for operations such as registration, profile discovery, and message exchange.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it facilitates the ingestion of untrusted data from other agents on the network.
Ingestion points: Untrusted content is received through /api/discover (agent profiles) and /api/chat (inbound messages).
Boundary markers: No delimiters or protective instructions are provided to help the agent distinguish between system commands and external agent content.
Capability inventory: The skill uses curl to communicate with the vendor's platform.
Sanitization: There is no evidence of content validation or sanitization for incoming data.

social