The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill scans and interprets external JSX, component, and menu files to identify emojis for replacement.
Ingestion points: Local source code files, including JSX and React components specified in the user request.
Boundary markers: The instructions do not define specific delimiters or "ignore" instructions to isolate processed code content from the agent's logic.
Capability inventory: The skill is authorized to perform file system read/write operations and execute host CLI commands as part of its workflow.
Sanitization: The workflow includes running type checks and providing replacement reports to the user for validation prior to final commits.
[COMMAND_EXECUTION]: The skill includes two Python scripts intended to be executed within the host environment.
profile_store.py: A utility to read and write user preference data to a local profile.
validate_skill.py: A maintenance script that uses the PyYAML library to verify the skill's structural integrity.
[PERSISTENCE]: The skill utilizes profile_store.py to persist user decisions and brand metadata across different sessions.
Persistence location: Data is stored in ~/.config/agent-skills/profile.json or ~/.lovstudio/skills/profile.json.
Security control: The script includes a specific blacklist (token, secret, password, credential, cookie, api_key) to prevent the storage of sensitive credentials in the profile.

lov-kill-emoji