The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external news emissions which could contain malicious instructions designed to influence the agent's behavior.
Ingestion points: Data is retrieved from the Ambush Streams service via the get_feed and list_emissions tools in SKILL.md.
Boundary markers: No delimiters or specific instructions to ignore embedded content are defined for the news payload processing.
Capability inventory: The skill can perform mutations including update_feed, route_feed_channel, and delete_feed.
Sanitization: There are no explicit instructions requiring the filtering or sanitization of emission content before it is processed by the agent.
[EXTERNAL_DOWNLOADS]: Connects to the official Ambush Streams MCP server at https://api.ambush.ai/mcp for stream management functionality.

manage-ambush-streams