The Agent Skills Directory

[SAFE]: The skill is composed entirely of instructional Markdown content and reference benchmarks. No executable code, scripts, or external dependencies are included.
[DATA_EXPOSURE]: The skill instructs the agent to read project-specific marketing context from local files such as .agents/product-marketing.md to inform its writing. These files are standard configuration paths for agentic workflows and do not target sensitive system credentials or private keys.
[INDIRECT_PROMPT_INJECTION]: The skill establishes a data ingestion surface by processing user-provided details about prospects and products. However, the risk is mitigated by the skill's lack of high-privilege capabilities (such as file writing or network operations) and the inclusion of specific writing principles and quality check frameworks that guide the agent's output generation.

cold-email