The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes a local sequenzy CLI to perform email marketing tasks. The execution patterns described are consistent with standard administrative operations for an email service.
[DATA_EXFILTRATION]: Network activity is restricted to legitimate service domains (api.sequenzy.com, sequenzy.com). The skill correctly instructs the use of environment variables for sensitive API keys rather than hardcoding credentials.
[EXTERNAL_DOWNLOADS]: No external scripts, remote payloads, or unverified third-party dependencies are downloaded or executed by the skill.
[PROMPT_INJECTION]: The instructions do not contain patterns intended to bypass AI safety guardrails or override core agent behavior.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data from various sources, presenting a potential surface for indirect injection.
Ingestion points: Files and strings passed to parameters such as --html-file, --blocks-file, --emails-file, and --filter-json in references/command-reference.md and references/use-cases.md.
Boundary markers: The skill includes explicit instructions in SKILL.md to "Validate IDs, recipient email, subject, template, or content input before issuing a mutation."
Capability inventory: Subprocess calls via the sequenzy CLI and network requests to the Sequenzy API for creating/sending emails.
Sanitization: The skill advises that generated content is draft material and must be reviewed by the user before deployment.

sequenzy-email-marketing