The Agent Skills Directory

[PROMPT_INJECTION]: No instructions were found that attempt to bypass AI safety filters, ignore prior rules, or extract internal system prompts. The workflow is strictly defined around the primary task of CTA design.
[DATA_EXFILTRATION]: The skill does not perform any network operations to external or untrusted domains. There is no evidence of attempts to access sensitive local files such as credentials, SSH keys, or environment variables.
[COMMAND_EXECUTION]: The skill utilizes allowed tools like 'AskUserQuestion' for interaction and 'Read' for accessing its own reference files. It does not attempt to execute arbitrary shell commands, perform privilege escalation (sudo), or establish persistence on the host system.
[REMOTE_CODE_EXECUTION]: There are no remote script downloads (curl | bash) or package installation commands. All logic is contained within the provided markdown instructions and local reference files.
[OBFUSCATION]: The skill's content is entirely in plain text. No hidden payloads, Base64 encoding, zero-width characters, or homoglyph substitutions were detected.
[INDIRECT_PROMPT_INJECTION]: The skill processes user input through an interview workflow. While this represents a data ingestion surface, the risk is mitigated by the structured nature of the recommended output and the lack of dangerous capabilities (like executing the user's input).

copywriting-cta