The Agent Skills Directory

[SAFE]: The skill functions as a technical guide and implementation reference. It contains no malicious instructions, prompt injections, or obfuscated code designed to compromise the agent or user data.
[EXTERNAL_DOWNLOADS]: The skill describes downloading language models via the AssetInventory API. These are documented as official system-managed assets from Apple and do not constitute a security risk.
[DATA_EXFILTRATION]: All transcription is explicitly described as on-device processing. There are no suspicious network operations or attempts to exfiltrate audio data or transcripts to external servers.
[COMMAND_EXECUTION]: The provided code snippets are Swift templates for iOS development and do not include any arbitrary command execution or shell injection vulnerabilities.

speech-analyzer