The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of untrusted data from external chat messages and space metadata.
Ingestion points: Data enters the agent context through methods such as spaces.messages (get/list) and spaces.list, which retrieve content authored by external users.
Boundary markers: The instructions do not define delimiters or specific safety instructions to isolate retrieved chat content from the agent's core instructions.
Capability inventory: The skill includes powerful capabilities such as spaces.delete, spaces.members management, and customEmojis modification across the gws CLI.
Sanitization: No sanitization, escaping, or validation logic is specified for processing the content retrieved from external API calls.

gws-chat