Security audit
Security checks for vulnerabilities and agentic risk
This is a disclosed marketing-operations workflow skill with no code, install scripts, hidden execution, or credential handling, though users should authorize any real publishing or email sends.
Install this when you want an agent to coordinate a full marketing workflow. Before allowing execution, confirm which social accounts, blog systems, email cohorts, customer data, and writeback locations it may use, and require review before real public posts or lifecycle sends.
The skill description is broadly phrased around running, auditing, improving, packaging, and documenting a full marketing operating system across many functions. That increases the chance of overbroad activation on loosely related marketing tasks, causing the agent to pull in an expansive multi-agent workflow when a narrower skill would be safer and more appropriate.
The invocation guidance says when to use the skill but does not define strong non-activation boundaries beyond a couple of examples. In practice, this can cause the agent to over-select a powerful orchestration skill for ordinary content or marketing requests, expanding scope, increasing unintended actions, and making downstream behavior harder to constrain.
No suspicious patterns detected.
