The Agent Skills Directory

[SAFE]: The skill consists exclusively of markdown instructions for text analysis and proofreading. It does not include any executable scripts, shell commands, or binary files.
[PROMPT_INJECTION]: The instructions are designed to guide the agent in text correction and do not contain any patterns intended to bypass safety filters, override system instructions, or extract system prompts.
[DATA_EXFILTRATION]: No sensitive file paths, credential patterns, or network exfiltration methods were detected. The skill operates locally within the agent's context using provided text.
[REMOTE_CODE_EXECUTION]: The skill does not perform any remote code downloads or dynamic execution of untrusted scripts. It references a local document references/sensitive-words.md for its knowledge base.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-provided novel text for proofreading, which is a standard ingestion point for this use case. There are no high-risk capabilities (such as command execution or network access) that could be exploited via malicious input in the processed text.

proofread