The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by demonstrating how to ingest and display user-supplied profile information (e.g., names and email addresses). If an agent uses these instructions to process user data without proper sanitization, it could be influenced by malicious content embedded in that data.
Ingestion points: User profile data is fetched using the useUser() hook in the ProfileScreen component.
Boundary markers: The instructions do not include specific delimiters or instructions to the agent to treat this data as untrusted content.
Capability inventory: The skill includes capabilities to update user metadata (user.update) and perform network requests (fetch) to backend services.
Sanitization: No sanitization or validation logic for the ingested user data is provided in the examples.
[EXTERNAL_DOWNLOADS]: The skill provides instructions to download and install standard, well-known libraries from the official Expo and Clerk registries.
Evidence: npx expo install @clerk/clerk-expo expo-secure-store in SKILL.md.
Context: These are industry-standard dependencies required for the skill's primary purpose of implementing authentication.

clerk-auth-expert