The Agent Skills Directory

[NO_CODE]: The skill consists entirely of markdown instructions and templates. It does not contain any executable scripts (Python, Node.js, Bash) or system commands.
[SAFE]: The skill uses a disable-model-invocation: true flag, which limits the AI to only using its internal knowledge to follow the provided instructions. There are no network requests, file system operations, or credential accesses detected.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-provided content (current outreach messages) to perform audits. However, it lacks any high-privilege capabilities (like sending emails or writing files) that could be abused if malicious instructions were embedded in that data. The 'Gotchas' section explicitly warns against automated sending.

outbound-optimizer