The Agent Skills Directory

[SAFE]: The skill consists of instructional markdown content focused on scientific editing. It does not contain any executable code, network requests, or attempts to access sensitive system resources.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external manuscripts, which constitutes a data ingestion surface. However, the risk is negligible as the instructions mandate a strict "source of truth" workflow, requiring the agent to distinguish between confirmed facts and assumptions, and no dangerous tools (such as network or file-write operations) are invoked or enabled within the skill.

manuscript-optimizer