The Agent Skills Directory

[SAFE]: The skill consists entirely of natural language instructions and guidelines for an AI to act as a UX writing assistant. It does not contain any executable scripts, binaries, or configuration files that could pose a technical security risk.
[PROMPT_INJECTION]: No malicious prompt injection patterns were detected. The use of terms like 'CRITICAL' and 'IMPORTANT' is within the context of emphasizing UX best practices and does not attempt to bypass AI safety filters or override system instructions.
[DATA_EXFILTRATION]: There are no commands or functions that attempt to access local files, environment variables, or credentials. No network operations (such as curl or wget) are present.
[REMOTE_CODE_EXECUTION]: The skill does not download or execute any external scripts or packages. It relies solely on the LLM's internal reasoning based on the provided guidelines.
[INDIRECT_PROMPT_INJECTION]: While the skill is designed to process user-provided interface text (an untrusted data source), it lacks any 'dangerous' capabilities such as file system access, network connectivity, or code execution. Consequently, even if a user were to provide text containing hidden instructions, the agent would only produce modified text, posing no risk to the host system.
Ingestion points: Processes user-provided UI copy via the target argument or conversation history.
Boundary markers: None present.
Capability inventory: No subprocess calls, file writes, or network operations.
Sanitization: Not applicable as the output is limited to text generation.

clarify