The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses a local Python utility scripts/profile_store.py to persist and retrieve user settings and brand guidelines from a shared JSON profile. The script includes protective logic to filter out sensitive keys (e.g., token, secret, password) and uses atomic write operations with restricted file permissions to ensure data integrity and security.
[INDIRECT_PROMPT_INJECTION]: The skill's primary function involves reading and refactoring external source code files, which serves as a potential ingestion point for untrusted content. The risk is mitigated by the skill's structured execution boundary and a multi-step workflow that requires analysis and planning before any modifications are committed.
[COMMAND_EXECUTION]: A validation script scripts/validate_skill.py is provided to perform structural and metadata linting of the skill package. This script requires the PyYAML package to be present in the execution environment to process YAML frontmatter and manifests.

lov-refactor-modular