The Agent Skills Directory

[SAFE]: The skill implements a structured 'Seven Sweeps Framework' for editing marketing copy. It focuses on linguistic improvements such as clarity, tone, and conversion optimization without requesting dangerous permissions or executing external code.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted user-provided marketing copy, creating a surface for potential indirect prompt injection attacks.
Ingestion points: The skill ingests marketing copy from the user and reads context from the local .claude/product-marketing-context.md file as specified in SKILL.md.
Boundary markers: The instructions do not specify the use of delimiters or specific warnings to ignore instructions embedded within the text to be edited.
Capability inventory: The skill instructions do not involve any shell commands, file system writes, or network activity; the agent's actions are limited to text analysis and generation.
Sanitization: There are no instructions for sanitizing or filtering input text.

copy-editing