The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-provided marketing copy for editing. While this creates a potential surface for indirect prompt injection if the source text contains malicious instructions, the risk is negligible because the skill does not possess capabilities for network access, file writing, or command execution that could be exploited.
[DATA_EXPOSURE]: The instructions direct the agent to read .claude/product-marketing-context.md if available. This is a standard pattern for retrieving project-specific context and does not target sensitive system files, credentials, or environment variables.
[NO_CODE]: The skill consists entirely of Markdown instructions and reference tables. It does not include any scripts, executable binaries, or external dependencies.

copy-editing