The Agent Skills Directory

[SAFE]: The skill consists of markdown-based documentation and reference materials for implementing security features in Apple platform applications. No malicious code, unauthorized data exfiltration, or persistence mechanisms were detected.
[PROMPT_INJECTION]: Documentation in skills/agentic-security.md includes strings like 'Ignore previous instructions' as examples of indirect prompt injection attacks. These are provided for threat modeling and defensive education purposes and do not represent instructions to the agent to override its behavior.
[EXTERNAL_DOWNLOADS]: The skill references official Apple documentation and well-known service repositories (e.g., github.com/apple/swift-crypto) for legitimate development purposes.
[COMMAND_EXECUTION]: Documentation in skills/code-signing-ref.md and others provide standard CLI commands (e.g., security, codesign, openssl) for common developer tasks like certificate management and signature verification.

axiom-security