The Agent Skills Directory

[SAFE]: The skill is comprised entirely of Markdown documentation (SKILL.md, references/*.md) and a JSON evaluation file (evals/evals.json). No scripts, binaries, or executable logic were found within the skill package.
[SAFE]: No network-enabled commands (such as curl or wget) or data exfiltration patterns were detected. The skill does not attempt to access sensitive system files or credentials.
[SAFE]: The instructions do not contain prompt injection attempts or obfuscated content. The logic for checking marketing context files is a standard convention for the target platform and does not pose a security threat.
[NO_CODE]: The skill provides instructional content and reference data for the AI agent to improve its email writing capabilities but does not include any functional code to be executed by the host system.

cold-email