The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a bundled Node.js script (scripts/learning-store.mjs) to manage practice records. The script uses standard Node.js modules to perform atomic JSON file writes, data validation, and passage similarity checks. It enforces strict file permissions (0o600 for files, 0o700 for directories) and uses directory-based locking to handle concurrent access.
[INDIRECT_PROMPT_INJECTION]: The skill processes IELTS reading materials sourced from external websites and publications, creating an attack surface for instructions embedded in third-party content. However, the skill instructions provide guidance for the agent to mitigate this risk.
Ingestion points: External IELTS passages and source materials processed in SKILL.md and practice.md.
Boundary markers: The instructions explicitly direct the agent to "verify the actual source and answer key online" and to "re-solve independently of the draft key" rather than trusting provided content blindly.
Capability inventory: The skill can execute the local learning-store.mjs script for data management and utilize PDF rendering tools for output.
Sanitization: The management script performs NFKC normalization and tokenization on passage text before performing duplicate detection or storage.

ielts-reading