The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-supplied content (articles, transcripts, and essays) to create learning tasks, creating a surface for indirect instructions.
Ingestion points: User-provided study materials defined in the Material Processing section of SKILL.md.
Boundary markers: Absent; the skill does not instruct the agent to use specific delimiters or ignore embedded instructions within the processed text.
Capability inventory: The skill can execute shell commands (via edge-tts), write HTML files, and perform network operations for Cloud TTS services.
Sanitization: No explicit sanitization or validation of input material is performed before task generation.
[DYNAMIC_EXECUTION]: The skill generates interactive HTML practice pages.
Evidence: The HTML Practice Page Mode in SKILL.md instructs the agent to create standalone files with inline JavaScript and CSS for UI controls, listen buttons, and progress tracking.
[EXTERNAL_DOWNLOADS]: The skill facilitates the installation of the edge-tts package.
Evidence: README.md and SKILL.md provide instructions for installing the package via pip to enable offline audio generation.
[COMMAND_EXECUTION]: The skill uses shell commands for environment setup and tool operation.
Evidence: Instructions include python -m pip install edge-tts and CLI usage for generating audio assets (e.g., python -m edge_tts --voice ...).

english-learning-coach