The Agent Skills Directory

[SAFE]: The skill is entirely instructional, providing guidance on how to write effective sales outreach emails. It does not contain any executable scripts, binaries, or command-line operations.
[DATA_EXPOSURE]: The instructions direct the agent to check for specific files such as .agents/product-marketing.md to gain context about the user's product. This is a standard context-gathering practice for marketing-focused agents and does not involve accessing sensitive system credentials or private keys.
[NO_CODE]: There are no code files (Python, JavaScript, etc.) or configuration files that trigger subprocesses or network requests included in this skill package.
[INDIRECT_PROMPT_INJECTION]: The skill has a minimal attack surface for indirect prompt injection. While it reads project-specific marketing files, it lacks the 'capability inventory' (such as subprocess execution or network writing) required to weaponize any instructions that might be found in those files.

cold-email