The Agent Skills Directory

[Prompt Injection] (SAFE): No attempts to override agent instructions or bypass safety filters were detected. The use of terms like 'CRITICAL' and 'HIGH' is limited to describing performance impact levels.
[Data Exposure & Exfiltration] (SAFE): No hardcoded credentials (API keys, secrets) or access to sensitive local file paths (e.g., .ssh, .aws) were found. No unauthorized network requests are present.
[Obfuscation] (SAFE): All content is provided in plain text markdown with no Base64 encoding, zero-width characters, or homoglyph-based evasion techniques.
[Unverifiable Dependencies & Remote Code Execution] (SAFE): The skill references standard, reputable libraries (e.g., Reanimated, Expo, LegendList). There are no remote script execution patterns (e.g., curl | bash) or suspicious dependency installations.
[Privilege Escalation] (SAFE): No commands related to privilege escalation, such as 'sudo' or modifications to system permissions, were identified.
[Persistence Mechanisms] (SAFE): No instructions to modify shell profiles, cron jobs, or system services for persistence were found.
[Metadata Poisoning] (SAFE): The metadata accurately identifies the author (vercel) and the purpose of the skill. No hidden instructions were found in the metadata fields.
[Indirect Prompt Injection] (SAFE): The skill serves as a static reference for coding best practices and does not ingest or process untrusted external data at runtime.
[Time-Delayed / Conditional Attacks] (SAFE): No logic was found that gates behavior based on date, time, or specific environment conditions.
[Dynamic Execution] (SAFE): The skill does not use 'eval()', 'exec()', or unsafe deserialization. Reanimated worklets are used appropriately within the framework's intended architecture.

react-native-skills