The Agent Skills Directory

[SAFE]: The skill is strictly for linguistic transformation and rephrasing. It defines formatting rules for English, Italian, and Spanish based on ISO 24495-1 and technical writing standards. No suspicious commands, network requests, or file access patterns were detected.
[SAFE]: The skill configuration includes disable-model-invocation: true, which acts as a strong security constraint by preventing the agent from executing any other tools or functions while the skill is active.
[INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it processes user-provided context and invocation arguments. 1. Ingestion points: User-provided argument-hint and conversation history. 2. Boundary markers: Absent. 3. Capability inventory: None. Tool invocation is explicitly disabled. 4. Sanitization: Absent, although the requirement to follow strict controlled-language dictionaries and grammar rules serves as a functional filter.

wait-what