The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill implements logic for fetching data from remote APIs and processing push notifications, which introduces a potential surface for indirect prompt injection if the incoming data contains malicious instructions.\n
Ingestion points: APIClient.shared.fetchLatestFeed() (SKILL.md) and didReceiveRemoteNotification (SKILL.md).\n
Boundary markers: The provided Swift code snippets do not include boundary markers or delimiters to differentiate between instructions and data.\n
Capability inventory: The skill has capabilities to write to the file system using FileManager.default.moveItem and perform network requests via URLSession.\n
Sanitization: There is no evidence of data validation or sanitization before processing the remote content.\n- [METADATA_POISONING]: The skill documentation repeatedly references "iOS 26+", a version that does not exist. This deceptive metadata misrepresents the skill's technical scope and compatibility.

background-processing