The Agent Skills Directory

[SAFE]: The skill provides instructions for a technical writing task. It uses standard file operations (read_file, patch, write_file) to process user-provided text. These operations are scoped to the intended purpose of rewriting documentation.
[SAFE]: References to external sites like asd-ste100.org and the author's GitHub repository (github.com/AminBlg/SimpleEnglish) are for documentation and reference purposes. These are well-known or author-related sources and do not present a security risk.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from user-specified files. While it includes instructions to ignore specific elements like code blocks and identifiers, there is a standard surface for indirect prompt injection if the processed content contains instructions that the agent might inadvertently follow. However, the skill's specific focus on grammatical rewriting significantly limits this risk.
Ingestion points: User-provided files accessed via read_file (documented in SKILL.md).
Boundary markers: The skill instructs the agent to preserve 'Untouchables' like code blocks and quoted errors, providing partial context boundaries.
Capability inventory: File reading (read_file), file writing (write_file), and targeted rewriting (patch).
Sanitization: No explicit sanitization of input text is defined; the skill relies on the agent's task-specific instructions.

simple-english