The Agent Skills Directory

[COMMAND_EXECUTION]: The skill is instructed to automatically execute git commit after verifying changes to the Obsidian vault to maintain version history of the captured notes.
[PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection as it ingests and processes untrusted user-provided language learning material to create or update local files.
Ingestion points: User-provided learning content (vocabulary, grammar, sentences, etc.) defined in the skill's main workflow.
Boundary markers: No explicit boundary markers or specific instructions to ignore embedded commands are present when handling user data.
Capability inventory: The skill has read/write access to a local file system path and the ability to execute shell commands (git commit) via a subprocess.
Sanitization: No sanitization, escaping, or validation of user-provided content is performed before the data is written to the file system.
[DATA_EXFILTRATION]: The skill accesses a hardcoded local directory path at /Users/kidow/Library/Mobile Documents/iCloud~md~obsidian/Documents/lexivault. This path is associated with the skill author's environment and is used as the storage location for the Obsidian vault. No network operations or external data transmissions were detected.

capture