The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill implements a speech recognition pipeline that ingests data from a microphone, which is an untrusted external source.
Ingestion points: The speech.transcriptions flow in SKILL.md collects text data from voice recognition.
Boundary markers: The provided code snippet does not implement boundary markers or instructions to ignore embedded commands within the transcribed text.
Capability inventory: The skill interacts with UI components via showPartialTranscript and a state management function commitTranscript.
Sanitization: There is no evidence of sanitization or validation of the transcribed text before it is processed or displayed.

speech