The Agent Skills Directory

[SAFE]: The skill is primarily a prompt-based template generator. It does not contain any executable code, shell commands, or external network requests.
[DATA_EXFILTRATION]: No data exfiltration risks detected. File operations are restricted to writing structured markdown files to local project memory directories (memory/influencer/ and memory/hot-cache.md).
[PROMPT_INJECTION]: The instructions do not contain patterns for bypassing safety filters or overriding system instructions. The 'Disclosure and usage rights' section actually reinforces compliance requirements for marketing content.
[EXTERNAL_DOWNLOADS]: No external dependencies or remote code patterns are present. The skill relies entirely on user input and local reference templates.
[INDIRECT_PROMPT_INJECTION]: While the skill ingests untrusted user input to generate briefs, the output is saved as a markdown document for human review. It does not pass this data to dangerous execution tools (e.g., shell or eval), minimizing the impact of potential indirect injections.

brief-generator